Violin 6000 Series Memory Array User's Guide - Diamond Point ...

iolin V MEMORY

Violin 6000 Series Memory Array User’s Guide For Release V6.0.0

Document Number: 535-0008-00 Rev 10 May 2013

LEGAL NOTICE Copyright © 2010-2013 Violin Memory, Inc. All rights reserved. Violin Memory, Violin Technologies, Violin, vSHARE, vCACHE, Flash Forward, and Violin and Design are trademarks, registered trademarks or service marks of Violin Memory, Inc. (“Violin”) in the United States and other countries. All other brands, product names, company names, trademarks, and service marks are the properties of their respective owners. This document and the associated software product are protected by copyright and international treaties, and are distributed under license from Violin, including restrictions on their use, copying, redistribution and reverse engineering. Unless otherwise agreed by Violin in writing, Violin’s standard end user license agreement shall apply, which may be reviewed at www.violin-memory.com/legal. No part of this document may be reproduced, adapted or translated without prior written permission of Violin, except as permitted under applicable copyright law. The associated software product may include, access or otherwise operate, interface or be delivered with third party software or other applications or copyrighted materials, which are copyrighted and licensed by Violin suppliers. Such third party materials and licenses are identified in this document and/or at www.violin-memory.com/legal. Violin assumes no responsibility for any typographical, technical or other error or omission in this document. Violin reserves the right to periodically change the information contained in this document, but Violin makes no commitment to provide any such changes, updates, enhancements or other additions in a timely manner or at all. The only warranties for Violin products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. THIS DOCUMENT (INCLUDING ANY EXAMPLES AND OTHER INFORMATION CONTAINED HEREIN) IS MADE AVAILABLE “AS IS” WITHOUT REPRESENTATION OR WARRANTY OF ANY KIND. VIOLIN MAKES NO REPRESENTATION OR WARRANTY IN THIS DOCUMENT REGARDING ANY ASSOCIATED SOFTWARE OR ANY OTHER VIOLIN OR THIRD PARTY HARDWARE, SOFTWARE OR OTHER PRODUCTS OR SERVICES REFERENCED HEREIN. TO THE FULLEST EXTENT PERMITTED BY LAW, VIOLIN (FOR ITSELF AND ITS LICENSORS AND OTHER THIRD PARTIES IDENTIFIED HEREIN) HEREBY DISCLAIMS ALL REPRESENTATIONS AND WARRANTIES, WHETHER EXPRESS OR IMPLIED, ORAL OR WRITTEN, WITH RESPECT TO THE FOREGOING, INCLUDING WITHOUT LIMITATION, ALL IMPLIED WARRANTIES OF TITLE, NON-INFRINGEMENT, QUIET ENJOYMENT, ACCURACY, INTEGRATION, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE. IN NO EVENT SHALL VIOLIN (OR ITS LICENSORS OR ANY OTHER THIRD PARTY IDENTIFIED HEREIN) BE LIABLE CONCERNING ANY USE OF THIS DOCUMENT, REGARDLESS OF THE FORM OF ANY CLAIM OR ACTION (WHETHER IN CONTRACT, NEGLIGENCE, STRICT LIABILITY OR OTHERWISE), FOR ANY DIRECT, INDIRECT, PUNITIVE, INCIDENTAL, RELIANCE, SPECIAL, EXEMPLARY OR CONSEQUENTIAL DAMAGES, INCLUDING WITHOUT LIMITATION, ANY LOSS OF DATA, LOSS OR INTERRUPTION OF USE, COST OF PROCURING SUBSTITUTE TECHNOLOGIES, GOODS OR SERVICES, OR LOSS OF BUSINESS, REVENUES, PROFITS OR GOODWILL, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Violin Memory, Inc. 685 Clyde Avenue Mountain View, CA 94043 USA

ii

Violin 6000 Series Memory Array User’s Guide

535-0008-00 Rev 10

Table of Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 CHAPTER 1. Violin 6000 Series Memory Array Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . New in This Release. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . About the Memory Array. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Violin 6000 Series Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7 8 8 9

Violin Intelligent Memory Modules (VIMMs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Violin Switched Memory (VXM). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 vRAID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 PCIe Switching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

Product Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Violin 6000 Series Platform Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 System LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 Power Supplies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16 Fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

Violin 6000 Platform Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Memory Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 Power Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 Array Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Memory Gateways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Network Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 vRAID Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Violin Intelligent Memory Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 VIMM Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 Flash-Based VIMMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 VIMM Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22 VIMM Slot Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

System Configuration Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Management Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Command Line Interface (CLI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 Violin Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 E-mail Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 Syslog Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 SNMP Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29 XML REST API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29

CHAPTER 2. Manage, Monitor, and Administer with the Violin Web Interface . . . . . . . . . . . . . . . . . . 31 Introduction to the Violin Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 User Interface Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32 System and Browser Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Abbreviations for Large Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34 Web Interface Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34

Logging In and Out of the Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Logging In to the Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35

535-0008-00 Rev 10


i

Logging Out of the Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Accessing and Optimizing the iPad Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Launching the iPad Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Optimizing the iPad Web Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Web Interface Task Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Home . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Manage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39 40 41 42

Home Dashboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Dashboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . System View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . LUN Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Front and Rear Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

43 48 51 54

Manage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Manage LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Manage Initiators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Manage Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Manage LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Data Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Thin Provision LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Manage Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Space Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Monitor Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitor LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitor LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitor Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

106 110 111 111

Administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Cluster Admin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Notification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tools. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

114 122 128 137 139

CHAPTER 3. Manage the Memory Array with the Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Introduction to the Memory Array Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 High-Level Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Web Interface Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Web Interface Help. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Configuring the Violin Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Printing Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

144 145 145 146 146

Logging In to and Out of the Array Web Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 Logging Out of the Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Performing Tasks Using the Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 Viewing Status of the Memory Array. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 Showing Status of the Memory Array Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viewing VIMM Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Focusing on VIMMs Managed by vRAID Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Focusing on Modules Managed by Power Controllers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

149 150 151 152

Monitoring the Memory Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 Administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Array Admin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Managing User Accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Notification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tools. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

154 158 163 168 170 171

Monitoring the Memory Array Using the XTree Browser Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 173 ii


535-0008-00 Rev 10

CHAPTER 4. vSHARE Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Understanding vSHARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 vSHARE System Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .176 Support for vSHARE InfiniBand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177 Target Ports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177

vSHARE Configuration Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 AIX Compatibility NACA Bit Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Configuring Storage Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Initializing Violin Memory Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .180 Viewing Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .182

Configuring Target Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 Supported Target Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183 Configuring Fibre Channel and InfiniBand Target Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183 Configuring iSCSI Target Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .184

Configuring Initiator Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Working with LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190 Creating LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .190 Viewing LUNs and Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192 Resizing LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192 Exporting LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .193

OpenVMS Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Creating a LUN and Specifying the UDID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .195 Setting and Removing a LUN UDID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .197

Working with LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Creating LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .197 Deleting LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .198 Modifying LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .198 Exporting a LUN Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .199 Displaying LUN Groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .200

CHAPTER 5. vSHARE Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Verifying Connections and Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Verifying Target Port Activity and Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .201 Verifying Container Activity and Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .203

Block Storage Media Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 vSHARE Block Storage Management Commands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Managing Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Supporting VMware VAAI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .212 Managing iSCSI Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .212 Managing Initiator Groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213 Managing LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .214 Managing Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .216

Managing Block Storage in the Violin Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Email Alert Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Enabling and Disabling Email Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .217 Customizing Email Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .218 Viewing Current Email Settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .218

Unified Logging Upload . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Uploading Log Files for All Nodes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .219 Uploading Log Files for Specific Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .219 Unified Logging Upload Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .219

CHAPTER 6. Manage the Memory Array with the CLI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Overview of the Command Line Interface (CLI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 CLI Shorthand Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .222 Getting Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .222 Tab Completion of Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .224 Command Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .225 535-0008-00 Rev 10


iii

Prompt and Response Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 Abbreviations for Large Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

Connecting to the Command Line Interface (CLI). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Logging into the Memory Array CLI via the Serial Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Configuration File Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230 Reverting without Saving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

Upgrading the System Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Downloading the Upgrade Image. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Upgrading All Firmware and Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Upgrading the Memory Gateway Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Upgrading the Memory Array Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

233 233 235 236

Reboot and Shutdown Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 Before You Proceed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rebooting the Memory Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shutting Down the Memory Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Powering Off and On Memory Array Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Powering Off and On Specific Memory Array Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Restarting the Memory Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

238 239 239 240 240 242

Rebalancing RAID Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 Discovering Unbalanced RAID Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Enabling Automatic Rebalancing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Enabling Incremental Rebalance of RAID Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Scheduling a RAID Group Rebalance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

243 244 244 245

CHAPTER 7. Data Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Managing Data Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 iSCSI Authenticated Sign-On . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Unidirectional CHAP Authentication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Bidirectional CHAP Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248

Setting Up iSCSI CHAP Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 CHAP Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Establishing a CHAP Secret. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Configuring Unidirectional iSCSI CHAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Configuring Bidirectional iSCSI CHAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

249 249 250 250

Removing CHAP Secrets from a Target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 Showing iSCSI CHAP Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252 CHAPTER 8. Data Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Violin Memory Data Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 Data Management Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254

Data Management Use Cases and Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Data Management Use Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Example Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

Best Practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Over Provisioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 Usage Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 Performance Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256

Optimizing Performance with ALUA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Setting Preferred Port and ALUA Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Viewing Preferred Port Settings and Sessions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 Viewing ALUA Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

Thin Provisioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 Thick LUNs versus Thin LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 Thin Provision LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 Creating Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 Deleting Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 Viewing Snapshot Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

iv


535-0008-00 Rev 10

Rolling Back the Contents of a Snapshot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267 Renaming Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .268 Modifying the Access to Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .269 Exporting Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .269 Protecting Snapshots from Automatic Deletion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .270

Snap Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270 Creating a Snap Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .270 Deleting a Snap Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .271 Showing a List of Snap Groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .271 Rolling Back the Contents of a Snap Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .272 Modifying a Snap Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .272 Changing the Name of a Snap Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .273 Exporting a Snap Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .273

Schedule Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 Creating a Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .274 Deleting a Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .275 Modifying Schedules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .276 Showing a List of Schedules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .279

Free Space Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 Monitoring Free and Consumed Space. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .279 Thresholds and the Reclamation Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .282 Rules for Reclaiming Free Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .283 Setting Usage Thresholds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .283

APPENDIX A. CLI Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 CLI Command Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 Key to Command Parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .288

CLI Option Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 Control User Inactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .289 Control Paging of CLI Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .290 Control Showing Hidden Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .290 Confirm Loss of Unsaved Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .290 Confirm Reboot or Halt Reload . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .290 Set Terminal Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .291 Clear Command History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .291 Display CLI Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .291 Display Terminal Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .291

General Configuration Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Verify Hardware Version and Capabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .292 View Installed Licenses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .294 Event Logging Configuration and Viewing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .294 User Accounts and Local Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .299 NTP, Clock and Time Zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .300 Event Notification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .302 Diagnostic Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .303 Configuration Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .305 Configuration File Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .306 Image Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309 Enable and Config Modes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .310 Web Proxy Settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 Xinetd Server Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 tcpdump Capture Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 NTB Commands (vMOS Data Management) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .312 Cluster Configuration and Show Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .313

Network Configuration Commands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314 Network Bond Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .315 Ten network bonding modes are supported: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .315 Name Resolution Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .315 Routing Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .316 535-0008-00 Rev 10


v

Administrative Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 donotuse Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

Email Alert Consolidation Commands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 Enable and Disable Email Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 Customize Email Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 View Current Email Settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318

Unified Logging Upload Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Address Resolution Protocol Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Add and Remove Static ARP Entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Clear Dynamic ARP Entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Display Contents of ARP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List Statically-Configured ARP Entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

319 320 320 320

Array Configuration and Show Commands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 Array Configuration Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 Array Show Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Array Balance Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322

VIMM Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 iSCSI Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 Show iSCSI Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Enable iSCSI Target Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Create an iSCSI Target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bind an iSCSI Target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

325 326 326 326

Security Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 iSCSI CHAP Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 RADIUS Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328 TACACS+ Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329

SCSI-3 ALUA Commands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330 Fibre Channel Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 Initiator Group Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 Create an Initiator Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Add Initiators to a Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Create a Nested Initiator Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Show Initiator Groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Show Nested Initiator Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Show Initiators on Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

332 332 332 332 333 333

LUN Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Create LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . View LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Set LUN Privileges and UDIDs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rename a LUN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Resize a LUN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Export LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . View LUN Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

333 334 335 335 336 336 337

LUN Group Commands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338 Creating LUN Groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deleting LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modifying LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Exporting a LUN Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Displaying LUN Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

338 338 338 339 340

Data Management Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341 Optimizing Performance with ALUA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Thin Provision LUNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Snap Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Schedule Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Free Space Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

341 344 346 350 353 356

Diagnostic and Display Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 Standard Show Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 Standard CLI Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361 vi


535-0008-00 Rev 10

Authentication Method and Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .362 SSH Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .362 Banner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .364 CLI Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .364 SNMP Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .365 Web Interface Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .366

APPENDIX B. Alarm Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369 Understanding Memory Array Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369 Interpreting Alarm LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .370 Understanding System Alarm Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .371

Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372 Equipment Service-Affecting Alarms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .372 Flash Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .372 Port Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .373 Power Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .373 RAID Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .374 System Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .375 Temperature Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .376 VIMM Service-Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .376

Non-Service Affecting Alarms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 Temperature Non-Service Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .378 Equipment Non-Service Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .378 Flash Non-Service Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .379 Port Non-Service Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .380 Power Non-Service Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .380 RAID Non-Service Affecting Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .380 VIMM Non-Service Affecting Alarms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .381 vRAID Controller Module (VCM) Alarms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .386

APPENDIX C. Violin Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 Violin Linux Drivers Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 Understanding Violin Linux Driver Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .388 Installation and Configuration Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .388

Installing Violin Linux Driver Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388 Kernel Development Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .389 GCC Compiler. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .389 Libaio Development Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .389

Installing Violin Linux Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 Installing Violin Linux Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .390 Uninstalling Violin Linux Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .393

Loading and Unloading Linux Drivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394 Loading Violin Linux Drivers Manually . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .394 Loading Violin Linux Drivers as Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .395 Unloading the Violin Linux Driver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .395

Installing Violin Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 Installing Violin Utilities on Linux Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .396

Configuring Partitions and Storage Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397 Creating Block Device Partitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .397 Enabling Host to Automatically Mount Partitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .400 Enabling Logical Volume Manager (LVM) to Recognize Direct-Attached Storage . . . . . . . . . . . . . . . . . . . .401

APPENDIX D. Violin Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403 Understanding Violin Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403 Installing Violin Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .403 Running Violin Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .403

Violin Utilities Reference. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405

535-0008-00 Rev 10


vii

varray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 vcounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406 veeprom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 vincident -a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410 vinfo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410 vinventory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412 vmesg. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 vpartial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 vring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415 vspeedtest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415 vstat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416 vvimms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419 vzero . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419 perf_test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420

APPENDIX E. Performance Tuning and Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423 Performance Tuning and Testing Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423 Flash-Layer Performance Tuning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424 Memory Array Flash 4kB Alignment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425

Network-Layer Performance Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427 Preliminary Network-Layer Checks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427 Performance Testing Using Perf_Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 428

APPENDIX F. SNMP Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433 About Violin Memory SNMP MIBs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benefits of Using a MIB Browser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SNMP Configuration on the Violin 6000 Series Memory Array. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Trap/Notify Event E-mail and Logged Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

433 434 435 436

Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438

Available Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438 Configuring and Testing Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 Enabling Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 Testing Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 Spare VIMMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450 Failed VIMMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450 PSU States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450 Temperatures: per VIMM and Chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451 Performance Stats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452 Fibre Channel (vSHARE) Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 Fibre Channel Performance Statistics (vSHARE) Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454

New Trap MIB Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454 APPENDIX G. Optimizing Connectivity for Initiator Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465 Optimizing Connectivity for Windows Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465 Windows Server 2008 R2 and Later . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465 Windows Server 2003 R2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466

Optimizing Connectivity for AIX Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466 System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Installing the Violin Driver for AIX MPIO. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Configuring an AIX Initiator Host for Optimum Connectivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discovering AIX Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Uninstalling the Violin Driver for AIX MPIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

viii


467 467 468 469 469

535-0008-00 Rev 10

System Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471 Memory Array System Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fibre Channel Adapter Card Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . InfiniBand Adapter Card Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PCIe Host Adapter Card Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ethernet/iSCSI Adapter Card Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

472 473 474 474 475

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477

535-0008-00 Rev 10


ix

x


535-0008-00 Rev 10

Preface

This preface outlines the organization of this book, describes document conventions, and provides information about additional resources.

• • • • •

Intended Audience on page 1 Document Organization on page 2 Reference Documents on page 3 Document Conventions on page 3 Contacting Violin Memory on page 5

Intended Audience This guide is intended for experienced systems administrators. Violin Memory assumes that you are experienced in installing and servicing high-performance storage systems. Contact Violin Memory Customer Support for any assistance with installing and servicing this system. See Contacting Violin Memory on page 5 for contact information.

535-0008-00 Rev 10


1

Document Organization This guide is organized into the following sections:

2

•

Chapter 1, Violin 6000 Series Memory Array Overview—Introduces the Violin 6000 Series Memory Array and its features, architecture, and specifications.

•

Chapter 2, Manage, Monitor, and Administer with the Violin Web Interface—Describes Web interface tasks specific to the Memory Gateways that are internal to the Violin 6000 Series Memory Array.

•

Chapter 3, Manage the Memory Array with the Web Interface—Describes using the Violin Web interface to complete the basic Memory Array monitoring and administrative tasks and the tasks specific to the Memory Gateways that are internal to the Violin 6000 Series Memory Array.

•

Chapter 4, vSHARE Configuration—Describes vSHARE, its features and functions, and the steps required to configure a Memory Gateway running vSHARE to manage block storage.

•

Chapter 5, vSHARE Management—Describes the management of a Memory Gateway configured for vSHARE block storage using the command line interface (CLI) and Violin Web Interface.

•

Chapter 6, Manage the Memory Array with the CLI—Provides detailed procedures for configuring and showing status of the Violin 6000 Series Memory Array and its internal modules using the Command Line Interface (CLI).

• •

Chapter 7, Data Security—Covers available options for enhancing data security.

•

Appendix A, CLI Reference—Describes the Violin 6000 Series Memory Array command line interface (CLI) and how to use it; describes the commands available in the Violin 6000 Series Memory Array.

•

Appendix B, Alarm Reference—Describes the various system alarms that may occur when using the Violin 6000 Series Memory Array.

•

Appendix C, Violin Drivers—Describes how to install, use and remove the Linux drivers on a host computer directly attached to the Violin 6000 Series Memory Array.

•

Appendix D, Violin Utilities—Describes the Violin 6000 Series Memory Array utility commands and how to use them.

•

Appendix E, Performance Tuning and Testing—Describes how to perform performance tuning and testing on the Violin 6000 Series Memory Array.

•

Appendix F, SNMP Traps—Describes using SNMP Traps with the Violin 6000 Series Memory Array.

•

Appendix G, Optimizing Connectivity for Initiator Hosts—Demonstrates ways in which you can optimize the connectivity of initiator (client-side) hosts with Violin Memory Arrays.

Chapter 8, Data Management—Covers data management features, procedures, and commands.


535-0008-00 Rev 10

Reference Documents In addition to this guide, the following Violin Memory documents comprise the documentation suite that will assist you with setting up, using and servicing the Violin 6000 Series Memory Array. These guides are available for download from the Violin Memory Support site at http://www.violinmemory.com/support/

This document...

Provides this information...

Release Notes

This document describes the new features, resolved issues, known limitations and software upgrade instructions for the current release.

Violin 6000 Series Memory Array Installation Guide

This guide provides instructions for installing the Violin 6000 Series Memory Array in an equipment rack and completing the system setup and configuration.

Violin 6000 Series Memory Array Slide Rail Installation Guide

This guide provides instructions for installing the Violin 6000 Series Memory Array slide rails in an equipment rack.

Violin 6000 Series Memory Array Cable Management Installation Guide

This guide provides instructions for installing the cable management arm on a Violin 6000 Series Memory Array. It is included in the Violin 6000 Series Memory Array shipping box.

Reference Documents

Document Conventions Safety Icons The table below summarizes warning, caution, and note icons used in this document and includes sample text. Safety Icons Icon

WARNING!

535-0008-00 Rev 10

Sample Text

WARNING! Only authorized, qualified, and trained personnel should attempt to work on this equipment.


3

Safety Icons (Continued) Icon

Sample Text

Caution:

Caution: Follow the listed safety precautions when working on the Violin 6000 Series Memory Array.

Note:

Note: Read through this entire chapter and plan your

installation according to your location before installing the equipment. The following procedures and the order in which they appear are general installation guidelines only.

Typographical Conventions The following typographic conventions are used in this guide:

Format

Meaning

Bold

User Interface text.

Italic

Provides emphasis and identifies variables and document titles.

Courier

Command names, examples, and output.

Courier bold

Input you must type exactly as shown.

Information for which you must supply a value.

[ ]

Optional command parameters are enclosed within square brackets.

|

Separates a set of command choices from which only one may be chosen.

{ }

Required command parameters that must be specified are enclosed within curly brackets.

Typographical Conventions

Security Violin Memory, Inc., cannot be responsible for unauthorized use of equipment and will not make allowance or credit for unauthorized use or access.

4


535-0008-00 Rev 10

Contacting Violin Memory To obtain additional information or technical support for Violin Memory products, contact us at: Phone: 1-855-VIOLIN-5 (1-855-846-5465) International: +1 650-396-1500 Extension 3 Web site: http://www.violin-memory.com Email: [email protected] When contacting Violin Memory Customer Support, please have the following information available:

• • •

Model and serial number of the system for which you are requesting support. Software version. A brief description of the problem.

535-0008-00 Rev 10


5

6


535-0008-00 Rev 10

CHAPTER 1

Violin 6000 Series Memory Array Overview

This chapter introduces the Violin 6000 Series Memory Array in the following sections:

• • • • • • • •

535-0008-00 Rev 10

New in This Release on page 8 About the Memory Array on page 8 Violin 6000 Series Architecture on page 9 Product Models on page 10 Violin 6000 Series Platform Overview on page 11 Violin 6000 Platform Components on page 17 System Configuration Options on page 27 Management Interfaces on page 27


7

Chapter 1: Violin 6000 Series Memory Array Overview

New in This Release The following features and enhancements are new in this release:

•

Data Management vMOS Data Management is introduced on Violin 6000 Series flash Memory Arrays with the release of vMOS-6. vMOS Data Management allows you to manage and protect critical data, as well as simplify backup and disaster recovery using the following features:

•

Thin provision LUNs for better utilization of storage

•

Snapshots of individual LUNs and LUN groups

•

Snapshot groups for management purposes

•

Schedules for automated snapshots

•

Manage and monitor provisioned and allocated space on the array

For information on Web interface procedures, see Manage LUN Groups on page 76. For indepth information on CLI procedures and commands, see CHAPTER 8, “Data Management” on page 253.

•

LUN groups This release introduces the ability to group LUNs for simplified administration and management. It also provides the ability to monitor a group of LUNs as a combined entity. For information on Web interface procedures, see Manage LUN Groups on page 76.For in-depth information on CLI procedures and commands, see Working with LUN Groups on page 197.

•

Nested Initiator Groups This release introduces enhanced support for nested initiator groups (igroups). For information on creating nested initiator groups, see Creating, Modifying, and Deleting Nested Initiator Groups on page 72 and Managing Nested Initiator Groups on page 213.

•

SCSI-3 ALUA This release introduces an enhancement for Asymmetric Active-Active ALUA for enhanced Data Management performance. For more information, see Data Management on page 253.

About the Memory Array The Violin 6000 is a purpose-built memory array designed from the ground up to maximize the performance and value of flash memory and deliver a high-bandwidth, high IOPS, low latency and cost-effective storage system. The architecture supports various configurations with no single point of failure. All active components may be configured with redundancy and the ability to field replace modules while the system is operational. Each system can be optionally configured with integrated and redundant memory gateways that provide SAN/LAN integration. The Memory Array has significant flexibility in its configuration.

8


535-0008-00 Rev 10

Violin 6000 Series Architecture

The primary operating roles for the Memory Array are:

•

SAN-attached Storage (vSHARE): connects to a storage area network (SAN) via the integrated or external vSHARE Memory Gateways using Fibre Channel, InfiniBand or iSCSI/ Ethernet. This configuration allows multiple hosts, including clustered applications, to share the same storage.

•

Direct-attached Storage: connects via PCI Express (PCIe) to one or two host computers running Linux and Windows. This configuration provides low latency and high bandwidth.

The Violin 6000 is ideal for any application that requires active processing or rapid access to large amounts of data, including applications for transaction processing (Oracle, DB2, SQL Server, Sybase), analytics and data warehousing, messaging systems (e-mail), active file storage and metadata (GPFS, Lustre).

Violin 6000 Series Architecture The Violin 6000 utilizes technologies (VXM switched memory network, vRAID algorithm) and hotswappable components to enable thousands of flash devices to operate efficiently as a flash memory array.

Violin Intelligent Memory Modules (VIMMs) VIMMs provide the flash capacity for the Violin 6000 Series Memory Array. Up to 64 VIMMs are supported in a single Memory Array, with available capacities of 256GB or 512GB per VIMM. Each VIMM includes a flash controller that manages the flash memory, including flash media errors and failures, flash wear, and the process of “grooming” for removing unused or obsolete memory blocks from the flash memory. These processes are largely performed in state machines for minimum latency and maximum IOPS. See Violin Intelligent Memory Modules on page 20 for more on the technology, types, configurations and slot locations of the system VIMMs.

Violin Switched Memory (VXM) One core technology of the Memory Array is a switched memory network that combines the best attributes of a high-performance interconnect including sub-µsec latency, redundancy, scalability, and high availability. VXM connects the VIMMs to the vRAID Controllers and provides a high bandwidth, highly reliable data path. Much like PCI Express (PCIe), VXM is designed for low-latency memory accesses. Unlike PCIe and other technologies, VXM was specifically developed for applications that must handle large data sets and hence large numbers of flash devices and modules interconnected in a highly redundant, serviceable manner. The VXM architecture is inherently redundant, hot-swappable, and scalable. VXM supports four times the module density of PCIe. Each VIMM has three connections—any two connections may fail without loss of connectivity. Similarly, any two modules in a network may fail without the loss of data. The modules can be replaced without powering down the system or impacting user data access.

535-0008-00 Rev 10


9


The Memory Array includes four vRAID Controllers. These act as the roots of the VXM networks and can be configured for complete redundancy.Three vRAID controllers out of four can fail without loss of connectivity to all of the VIMMs. These vRAID controllers can also be hot-swapped as required.

vRAID The Memory Array employs a vRAID technology to optimize key flash storage attributes including latency under load, bandwidth, storage efficiency and reliability. vRAID provides full RAID data protection across the array while simultaneously guaranteeing that any read will not be blocked by an erase providing for spike-free latency under load and sustained performance. Unlike existing RAID 5 and RAID 6 solutions, which were designed for hard disk drives and rely on Read-Modify-Write operations that are unsuited to flash memory, vRAID was specifically designed to enhance flash storage systems. It uses an efficient 4+1 Parity model and writes any 4kB block without a requirement for Read-Modify-Write.

PCIe Switching The core of the Memory Array includes redundant PCIe switches that connect the vRAID controllers to the external PCIe ports or the integrated Memory Gateways. PCIe switches are high bandwidth and low latency and are an ideal complement to the VXM technology used for flash aggregation. Unlike the PCIe used in standard servers, the PCIe switches are implemented to support hot-swap of all modules for full redundancy. Each vRAID Controller is dual-homed to two different PCIe switches.

Product Models This Violin 6000 Memory Array is available in thee following models.

Model #

Configuration

6606

24 SLC VIMMs

6212

24 MLC VIMMs

6611

44 SLC VIMMs

6222

44 MLC VIMMs

6616

64 SLC VIMMs

6232

64 MLC VIMMs

Table 1.1 Violin 6000 Series Memory Array Models

10


535-0008-00 Rev 10

Violin 6000 Series Platform Overview

Violin 6000 Series Platform Overview This section provides an overview of the Violin 6000 Series flash Memory Array. See the Violin 6000 Series Memory Array Service Guide if you need to replace any of the system modules.

Chassis The Memory Array is a 3U high chassis and mounts on adjustable sliding rails into equipment racks from 19" to 36" deep.

Figure 1.1 Memory Array

The chassis itself mounts on slide-out rails. The lid can be removed without taking the system out of service.

Caution: To avoid overheating and the generation of system alarms, the lid must not be off of the system for more than 5 minutes when the ambient temperature is 35 °C (95 °F) and above.

•

The front panel LEDs and the Violin Web interface can indicate which units are faulty and require replacement

• •

All system components are hot swappable The power supplies are serviceable from the rear

535-0008-00 Rev 10


11


Top View Memory Array components are hot-swappable from the top of the chassis.

Network Interfaces

A

Power Controllers

B VIMMs

Internal Memory Gateways

B A

B

Array Controllers

B

A

C

D vRAID Controllers

A

A1

B1

C1 Fans

A0

B0

C0

Figure 1.2 Memory Array: Top View

Seven components are accessible from the top of the chassis:

12

Fans

Six fans are arranged in two rows of three each. Each fan may be replaced independently of others.

Array Controllers

The Array Controllers provide PCIe interfaces, switching, and management services including Ethernet and serial CLIs.


Internal Memory Gateways are an optional feature of the Memory Array. Two Memory Gateways are installed in each Violin 6000 Series system.

Power Controllers

The redundant Power Controllers provide power to all controllers, modules and fans. Each slot or socket can be individually powered up or down to simplify remote control.


535-0008-00 Rev 10


Network Interfaces

The network interfaces provide the physical connectivity to the SAN or LAN via Fibre Channel, iSCSI or InfiniBand Host Channel Adapter (HCA) cards.

VIMMs

The Violin 6000 manages up to 64 VIMMs, all redundantly connected via sub-µsec paths.

vRAID Controllers

The Memory Array has four vRAID controllers for redundancy and additional bandwidth.

Front View The front panel contains two USB ports (A and B), the ID button, and six LEDs (AC-A, AC-B, System Status, Port/Interface Status, VIMM Status, ID).

USB Port A

USB Port B

AC-A LED AC-B LED System Status

ID LED/Button Port/Interface Status VIMM Status

Figure 1.3 Memory Array: Front View

The USB ports are used for performing software upgrades. See Upgrading the Memory Array Firmware on page 236. For more information on LEDs, including the ID LED/button, see System LEDs on page 15.

535-0008-00 Rev 10


13


Rear View

PCIe Gen2 Ports

B

Network Interface Cards and Ports

A A B

1

C

D A B

2

B

A

D

C

B

A

1

1

1

1

n D

n C

n B

n A

A

B

A B

ID A B

ID LED

Power Supply A

Power Supply B

Ground

Ethernet Ports

Serial Console Ports

Figure 1.4 Memory Array: Rear View

Access to the system is available through the following interface ports: Array Controller Serial Console Ports

Serial console interface. Accesses the command-line interface (CLI) on the system.

Array Controller Ethernet Ports

RJ-45 10/100/1000 Ethernet Interface. Accesses the CLI on the system via Telnet or Secure Shell (SSH) or other management protocols.

PCIe Gen2 Ports (Two Per Array Controller)

Each Array Controller may be connected via single or dual PCIe x8 connections. Each PCIe interface may operate as PCIe x8 Gen2 or Gen1. The system will auto-negotiate to PCIe x4 if required. See the Violin 6000 Series Memory Array Installation Guide for more information on direct-attached storage configurations.

Network Interface Cards and Ports

The Network Interfaces provide the physical connectivity to the SAN or LAN via up to eight 8Gbit Fibre Channel, InfiniBand or up to eight 10GbE ports. Memory Gateway A supports slots A and B; Memory Gateway B supports slots C and D. See the Violin 6000 Series Memory Array Installation Guide for more information on SAN-attached storage configurations.

Power Supplies

14

Each power supply has two LEDs. The bottom LED is reserved for alarms. The power supply must be replaced if the bottom LED is solid amber. Blinking amber indicates overcurrent. Power supplies are also hot-serviceable. For more information on power supplies, see Power Supplies on page 16.


535-0008-00 Rev 10


The PCIe and SAN/LAN connections may be removed and reinstalled without rebooting the system. Each interface port on the rear of the system has an associated LED indicating if it is in use or not.

System LEDs The Violin 6000 is designed for highly reliable data center operation featuring hot-swappable modules and no single point of failure for high availability. The VIMMs, controllers, and fans are hotserviceable from the top. The System Status and VIMM Status LEDs alert you of system and VIMMspecific alarms respectively. Service LEDs indicate which modules are faulty and enable you to replace the modules with no downtime. Six system LEDs are located on the front of the Memory Array. Front LEDs Six LEDs (AC-A, AC-B, System Status, VIMM Status, and Interface Status) are displayed on the front of the chassis.

AC-A

AC-B

Interface Status

ID LED Button

System Status

VIMM Status

Figure 1.5 Front Status LEDs

LED

LED State

Meaning

AC-A (USB0)

Green (solid)

Array Controller A is the master controller.

AC-B (USB1)

Green (solid)

Array Controller B is the master controller.

System Status

Green (solid)

The following system components are operating normally: VIMM, internal Memory Gateway, HBA, vRAID Controller Module, Array Controller Module, Fan, Front Panel Module, Power Controller Module.

Amber (solid)

An alarm has been raised on at least one of the following system components: VIMM, internal Memory Gateway, HBA, vRAID Controller Module, Array Controller Module, Fan, Front Panel Module, Power Controller Module. A system component is not powered or not present.

Grey (solid)

The system is off or starting up.

Table 1.2 Front LED Descriptions

535-0008-00 Rev 10


15


LED

LED State

VIMM Status

Meaning

Green (solid)

All VIMMs are functioning correctly.

Amber (solid)

An alarm has been raised for at least one VIMM. A VIMM is not powered or not present.

Interface Status Green (solid)

This LED is not currently functional.

ID

Identifies the Memory Array in a rack when the ID button is pressed or when enabled by a remote operator.

Blue (solid)

Table 1.2 Front LED Descriptions (continued)

Active system alarms may be viewed by running the show alarms command on the Memory Array. See Alarm Reference on page 369 for more information.

Power Supplies The chassis contains two power supplies that provide an AC power source for the Memory Array.

B

A

Power Supply B

Power Supply A

Figure 1.6 Power Supplies

The 90% efficient power supplies convert the incoming AC power from individual AC feeds to DC power used by the system. These 2000-watt c20 receptacles are redundant in their configuration. The power supplies are hot-swappable from the rear of the chassis.

16


535-0008-00 Rev 10

Violin 6000 Platform Components

Fans Six fans are located in the front of the chassis and provide forced airflow ventilation for the Memory Array. When the system is first powered on, the fans come up at high speed. Built-in temperature sensors automatically lower the fan speed if lower temperature conditions are detected.

Figure 1.7 Fan Module

The fans are arranged in two rows of three each. Every fan may be replaced independently of one another. Multiple fans may fail and the system will remain operational. See Figure 1.2, “Memory Array: Top View,” on page 12 for fan layout and numbering. Multiple temperature sensors within the system provide indication, alarm and automated system control functions. The fans switch to full speed in the event of overheating conditions, high ambient temperature and other fault conditions, including failure of a VIMM, vRAID controller module or the fans themselves. There is roughly a 20-25°C temperature rise inside the system when air is flowing freely in and out of the system. Removal of a fan appears like a fault condition, causing all fans to switch to full speed until the removed fan is replaced.

Caution: The Memory Array is designed to tolerate a single fan failure and work with inlet temperatures between 0°C and 35°C. Unobstructed airflow is expected. Any obstruction to airflow causes unnecessary temperature rise of components inside the system. The fans operate in high-speed mode when the inlet temperature rises above 30°C.

The fans plug into connectors inside the front of the chassis. For more information, see “Replacing a Fan” in the Violin 6000 Series Memory Array Service Guide.

Violin 6000 Platform Components The Violin 6000 is designed to deliver high system availability through a variety of fault tolerance techniques that both reduce system outages and minimize the time and expense of operation. Key approaches include no single point of failure, redundancy of modules, fail-in-place for flash memory components and hot-swappable modules.

535-0008-00 Rev 10


17


The vRAID Controllers, Array Controllers, Power Controllers, internal Memory Gateways, and network interface cards can all be provisioned redundantly. Any single failure is automatically handled until the faulty unit is replaced.

Memory Gateway

Fan

VIMM

Array Controller

vRAID Controller

Power Controller Network Interface

Power Supply

Figure 1.8 Memory Array Components

All active components are hot-swappable and can be replaced without powering down the system.

Memory Plane The base system has a set of boards that act as the memory plane and provide interconnection between all modules and controllers. The memory plane is passive to ensure an extremely high mean time between failures.

Power Controllers The redundant Power Controllers provide power to all controllers, modules and fans. Each slot or socket can be individually powered up or down to simplify remote control. The Power Controllers are designed so that no single failure will impact the ability of the system to provide storage services.

18


535-0008-00 Rev 10


Array Controllers The Array Controllers provide PCI Express (PCIe) interfaces, switching and management services. These include serial management of the system including power supplies, fans and memory infrastructure. Command line, Web, e-mail, Syslog and SNMP interfaces are provided. ACM Connection for vMOS-6 vMOS-6 requires that the ACMs be externally connected to each other to interconnect the internal Memory Gateways. This cable connection enables a Non-Transparent Bridge (NTB), which allows for high-bandwidth, low latency inter-cluster coordination between the two Memory Gateways for effective management of thin provisioned LUNs and snapshots. For more information, see the Violin 6000 Series Memory Array Installation Guide for Release V6.0.0.

Memory Gateways The internal Memory Gateways are integral components of the Memory Array, as they manage connectivity to SAN systems. Block storage (vSHARE) capabilities are provided. Violin Memory Gateways with vSHARE provide value-added functionality such as flexible Logical Unit Number (LUN) provisioning, LUN masking and reservations, and full-speed LUNs. Each LUN can use the full IOPS/bandwidth of the Violin 6000 platform if the other LUNs are idle.

Network Interfaces The network interfaces provide the physical connectivity to the SAN or LAN. Eight 8Gbit Fibre Channel, InfiniBand or eight 10GbE ports are available via four separate Network Interfaces.

vRAID Controllers A vRAID Controller Module (VCM) manages the Violin Switched Memory Network (VXM) of VIMMs and implements the vRAID algorithms. vRAID Controllers enable the VIMMs to be hot swapped and automatically rebuild RAID groups on the failure of a single VIMM. The Memory Array has four vRAID Controllers for redundancy and additional bandwidth.

535-0008-00 Rev 10


19


Violin Intelligent Memory Modules The RAID-protected Violin Intelligent Memory Modules (VIMMs) provide the flash memory and controllers required for high-performance flash storage.

Figure 1.9 VIMMs

Data is written to VIMMs using the logical block address. The flash controller on the VIMM assigns the logical block to a physical page within the flash of the VIMM. Metadata is used to map between logical addresses and physical addresses. Each VIMM includes:

• • • •

A high-performance flash memory controller Management processor DRAM (e.g., 3GB) for metadata NAND flash (e.g., 512GiB) for storage

Violin Memory Arrays achieve high sustained performance by distributing a complex task called “grooming” to each VIMM and performing this task in hardware. Grooming is the process by which flash with stale data is recovered and the flash memory is prepared to be used for new data. This process involves reading a flash address space (block), writing the still-current data to a new flash address space, and then erasing the previous flash address space. Erasing, reading, and writing is a time- and resource-consuming task, which the VIMM is optimized to complete efficiently.

20


535-0008-00 Rev 10


VIMMs are designed to enable the scalability of large arrays of flash memory. The advantages of this architecture are:

• • • • • •

Grooming on each VIMM for sustained write performance Low latency access to DRAM metadata and flash memory Safe access and local storage of metadata for fault recovery Integrated monitoring and management of flash memory health—bad blocks, bad die Distributed ECC correction for maximum bandwidth Hot-swap and redundancy management

Unlike most solid-state drives and PCIe cards, a failed flash device does not cause a VIMM to lose data or be taken out of service. ECC and RAID protection are used to protect the data from errors. Data is read from other VIMMs, rebuilt using the RAID algorithm, and written back to the failed VIMM. This VIMM writes the same data to a different flash device, avoiding the need to be replaced. The Violin 6000 also allows the VIMM to be replaced without powering the system down or disrupting access to user data. Each VIMM can be independently powered down and powered up. A fully populated system contains 48 data VIMMs, 12 parity VIMMs and four spare VIMMs for failure protection. All memory modules are hot-swappable with automated rerouting that will not interrupt data service. The normal operating temperature range of VIMMs is 40-69°C. As air flows into the Memory Array, the first row of VIMMs preheats air passing over the second row of VIMMs, which then preheats air flowing over the third row of VIMMs. This results in the third row of VIMMs usually operating hotter than the first two rows. VIMMs in this third row typically record higher temperatures. In the case of a module failure where the module must be replaced, the Memory Array does not have to be powered down. See the Violin 6000 Series Memory Array Service Guide for VIMM replacement instructions.

VIMM Types The Memory Array performance depends on the type of memory used and the number of modules. VIMMs are available in the memory types listed in the following table.

VIMM Type

VIMM Capacity

Usable Capacity Per 64 VIMMs

SLC

256 GiB

8 TiB at 65% format level

MLC

512 GiB

20.7 TiB at 84% format level

Table 1.3 Available VIMM Types

Flash-Based VIMMs Flash systems store up to 32 TiB of raw flash storage per system. The amount of usable memory in a system is the number of VIMMs used for data storage, which is four per RAID group (up to 64 VIMMs), multiplied by the amount of memory per module, multiplied by the usable percentage for which the system is formatted. For example, in a fully loaded system with 64 VIMMs, the usable memory is 48 VIMMs multiplied by 512 GiB VIMMs, multiplied by the usable percentage, such as 84%, which is 20.7 TiB.

535-0008-00 Rev 10


21


VIMM Configurations The Memory Array contains up to 64 flash-based VIMMs, which operate in the VIMM configurations listed in the following table.

VIMMs

Active

Spares

24

20

4

44

40

4

64

60

4

Table 1.4 VIMM Configurations

vRAID Failover Protection All configurations of the Violin 6000 Series flash Memory Array contain four vRAID Controller Modules (VCMs): A, B, C and D. In the unlikely event of a failed VCM, its VIMMs are automatically assigned to another VCM in the system. RAID groups for the three VIMM configurations are as follows:

• • •

24-VIMM configuration: one RAID group per VCM. 44-VIMM configuration: two RAID groups per VCM. 64-VIMM configuration: three RAID groups per VCM.

Any outstanding writes are committed to flash on the VIMMs, with no loss of data. Once the failed VCM is replaced and fully booted, its original VIMMs return (“failback”) to the replacement VCM.

Note: In a 24-VIMM configuration, VIMMs belonging to VCM A or B are available to either A or B, but not C and D (and vice versa). In a 44- and 64-VIMM configuration, the VIMMs from the failed VCM can failover to any of the three other VCMs.

If a RAID rebuild is in progress when a VCM fails, the rebuild process stops and then restarts once the VCM failover process completes. The System Status LED on the front of the Violin 6000 is amber if a VCM or other system component is faulty. See the Violin 6000 Series Memory Array Service Guide for replacement procedures. Utilizing Floating Spares Each Violin 6000 Memory Array model has four spare VIMMs. Two spares are available and ready in the event of a VIMM failure within another RAID group. The spare VIMMs available to VCMs A and B reside in slots 3 and 6; the spare VIMMs available to VCMs C and D reside in slots 13 and 16. See VIMM Slot Locations on page 23 for more information, along with a top-level view of where the spares reside in each configuration. Each VCM can support up to two VIMM failures in a single RAID group without loss protection from a third failure.

22


535-0008-00 Rev 10


The VIMM Status LED on the front of the Memory Array is amber when an alarm has been raised on a VIMM. Failed VIMMs should be replaced with new spares as soon as possible. See the Violin 6000 Series Memory Array Service Guide for replacement procedures.

VIMM Slot Locations The Memory Array is available in three Flash VIMM configurations. Table 1.5 lists the VIMM slot numbers used for the configuration. Use the information in the table when adding, removing, or replacing a non-working VIMM to ensure that the VIMMs are still in their correct positions. Figure 1.10 Figure 1.11 and Figure 1.12 show the VIMM slot locations for the three available configurations. Numbers are printed inside the chassis below each VIMM slot for easy identification when removing VIMMs and as a guide for the different configurations. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide for information on replacing an errored VIMM.

Number of VIMMs

Slot Numbers

24

0, 1, 3, 6, 8, 9, 10, 11, 13, 16, 18, 19, 43, 44, 45, 46, 47, 48, 55, 56, 57, 58, 59, 60

44

0-19 and 40-63

64

All slots (0-63)

Spare Locations

Slots 3, 6, 13, 16

Table 1.5 VIMM Slot Locations

All Violin 6000 Series Memory Array models contain four spare VIMMs.

535-0008-00 Rev 10


23


Figure 1.10 VIMM Slot Locations: 64-VIMM System

24


535-0008-00 Rev 10



535-0008-00 Rev 10


25



26


535-0008-00 Rev 10

System Configuration Options

System Configuration Options The table below shows the various Memory Array configurations.

Item

Size / Quantity

VIMM Types

256 GB SLC, or 512 GB flash (MLC)

Number of VIMMs

24, 44 or 64

Number of vRAID Controllers

4

Number of Array Controllers

2

Number of Power Controllers

2

Direct-attach PCIe Ports

1, 2 or 4


2

Network Interfaces

8 ports (Fibre Channel, 10GbE or InfiniBand)

Table 1.6 Memory Array Configurations

See the Violin 6000 Series Memory Array Installation Guide for diagrams of the configurations supported by Memory Arrays, along with external and internal Memory Gateways.

Management Interfaces The Memory Array is accessible through physical port interfaces on the rear of the chassis and can be accessed remotely using a number of other interfaces, all briefly described in the following table.

Interface

Usage

Command Line (CLI) Configuration, maintenance and troubleshooting Web

Monitoring the system, configuring LUNs and caching, etc.

SNMP

Monitoring system health in a consolidated management system

Syslog

Consolidating and correlating events from many systems

E-mail

Reporting system status on a daily basis and providing alerts on major events

XML REST API

Enables management systems (e.g., provisioning system) to automate functions and directly control the Arrays

Table 1.7 Management Interfaces

535-0008-00 Rev 10


27


Command Line Interface (CLI) The command-line interface (CLI) is the default management interface. It is accessible via:

• •

the rear serial console ports on the Array Controllers the rear Ethernet ports on the Array Controllers (Telnet/SSH)

The CLI provides a full set of commands and information for configuring and operating the platform. It uses a Cisco-style CLI interface with command completion and integrated help. For more information, see Manage the Memory Array with the CLI on page 221 and CLI Reference on page 287.

Violin Web Interface The Violin Web interface provides an intuitive graphical interface for performing common configuration and operational activities. Examples of these activities include:

• • • •

managing the Memory Array monitoring system status monitoring system usage viewing the front and rear panels

The most common browsers (Firefox, Internet Explorer, Safari, Chrome) are supported. For more information, see Supported Web Browsers on page 33.

E-mail Interface The e-mail interface allows reports and alerts to be sent through the local e-mail server to both local administrators and Violin Memory Customer Support (or its partners). These reports can be sent daily or when specific events occur that require attention. For more information, see Notification on page 168 or Event Notification on page 302.

Syslog Interface Logging of events via Syslog is supported. This includes all CLI and configuration events as well as any internal warnings or alarms. Using an external Syslog server, such as Splunk, allows syslog events from many Violin Memory Arrays and hosts to be consolidated and correlated. See Event Logging Configuration and Viewing on page 294 for more information.

28


535-0008-00 Rev 10

Management Interfaces

SNMP Traps Simple Network Management Protocol (SNMP) enabled management systems can use SNMP to track the status of the Memory Array. Both SNMP Traps and SNMP polling can be performed via a generic SNMPv2 interface. The Memory Array Management Information Base (MIB) is available for integration into 3rd party management systems such as HP Network Node Manager and IBM Tivoli. For more information, see SNMP Traps on page 433.

XML REST API An XML API built on a RESTful HTTP interface is supported. This API is well suited for use by systems that automate specific tasks such as provisioning. Most configuration and monitoring capabilities provided by the CLI are also available by the XML API. See Monitoring the Memory Array Using the XTree Browser Interface on page 173 for more information.

535-0008-00 Rev 10


29


30


535-0008-00 Rev 10

CHAPTER 2

Manage, Monitor, and Administer with the Violin Web Interface

This chapter introduces the Violin Web Interface and covers the following topics:

• • • • • • • •

Introduction to the Violin Web Interface on page 31 Logging In and Out of the Web Interface on page 34 Accessing and Optimizing the iPad Web Interface on page 36 Web Interface Task Maps on page 39 Home Dashboard on page 43 Manage on page 58 Monitor on page 105 Administration on page 113

Introduction to the Violin Web Interface The Violin Web Interface is a Web-based graphical user interface (GUI) that you can access by connecting to the master Memory Gateway. This section covers the following topics:

• • • •

535-0008-00 Rev 10

User Interface Overview on page 32 System and Browser Requirements on page 33 Abbreviations for Large Numbers on page 34 Web Interface Help on page 34


31

Chapter 2 - Manage, Monitor, and Administer with the Violin Web Interface

User Interface Overview The streamlined Violin Web Interface includes the following interactive components, as shown in Figure 2.1:

•

Main Menu Bar—Displays clickable icons for the major functional categories: Home, Manage, Monitor, and Administration.

•

Function Menu Bar—Changes to display the options for the currently active category. Click an icon to display the associated page and submenu options.

• •

Online Help—Brings up context-sensitive help for the currently active page.

• •

Copyright—Displays the current copyright information.

•

Software Version—Displays the version of the software currently running on the Memory Gateways.

Session—Shows the user name of the current user at the right, and exits the session when clicked. Board Status Tab—Expands to show board status information when clicked. Click the tab again to hide the board status page.

Figure 2.1 Violin Web Interface Overview Main Menu Bar

Copyright

32

Function Menu Bar

Board Status Tab


Online Help

Session

Software Version

535-0008-00 Rev 10

Introduction to the Violin Web Interface

System and Browser Requirements This section provides information on the Violin Web Interface requirements for browsers, display resolution, Adobe Flash Player, JavaScript, and cookies.

Supported Web Browsers The following table lists the supported browsers and versions for the Violin Web Interface. Operating System

Linux

Supported Browsers

Mozilla Firefox 4 and above Google Chrome 11 and above

Mac

Mozilla Firefox 4 and above Apple Safari 4 and above Google Chrome 11 and above

Windows

Windows Internet Explorer 9 and above Mozilla Firefox 4 and above Apple Safari 4 and above Google Chrome 11 and above

Table 2.1 Supported Web Browsers

Note: For Windows Internet Explorer, Compatibility Mode should be turned OFF.

The following table lists the recommended software and settings for optimum performance of the Violin Web Interface. Recommendation

Description

Display resolution

The minimum recommended display resolution for the Violin Web Interface is 1024 by 768 pixels.

JavaScript

JavaScript must be enabled for the Violin Web Interface.

Adobe Flash Player

Adobe Flash Player version 8 (or above) is required for viewing charts in the Violin Web Interface.

Cookies

Enable cookies for login and session management. Cookies are also used to store dashboard page settings.

Table 2.2 Recommended for the Violin Web Interface

535-0008-00 Rev 10


33


Abbreviations for Large Numbers The following abbreviations are used for large numbers throughout the Violin Web Interface: Abbreviation

Meaning

B

bytes

KB

kilobytes

MB

megabytes

GB

gigabytes

TB

terabytes

PB

petabytes

Table 2.3 Large Number Abbreviations

For E (exabytes), Z (zettabytes), and Y (yottabytes), a single letter omitting the “B” is used for units other than bytes, or to save space in a display.

Web Interface Help There are two types of available help:

•

Context-sensitive online help for any page in the Web Interface—Click the Page Help icon in the upper right corner of the Web Interface window.

•

A full PDF version of this guide—Select Administration > Information > Documentation and click the PDF link.

For more information on available Help, see Information on page 140.

Logging In and Out of the Web Interface This section demonstrates how to access the Violin Web Interface and then log in to a Memory Gateway. The role assigned to your user account specifies the privileges you have and the tasks you can perform after you log in. The pages that are available to you depend on your user role. Role

Description

Admin

The Admin user can access the Web Interface pages for managing, monitoring, and administrating.

Monitor

Users with monitor privileges can only access pages for monitoring overall status and media status.

Table 2.4 User Roles

34


535-0008-00 Rev 10

Logging In and Out of the Web Interface

Logging In to the Web Interface You access the Violin Web Interface through a Web browser. For more information, see Supported Web Browsers on page 33. To access the Violin Web Interface and log in, do the following: 1. Open a Web browser, as described in Supported Web Browsers on page 33. 2. In the browser URL field, enter one of the following:

•

http://

•

http://

•

http://

•

http://

Note: If you specify the IP address or hostname of a cluster node (other than the master), the connection is automatically redirected to the master Memory Gateway.

The Web Interface log in page appears. 3. Enter a valid username and password in the text fields of the Login page, as shown in the

following example.

4. Click Login. The Dashboard page appears. 5. Continue with Customizing the Dashboard on page 44.

535-0008-00 Rev 10


35


Logging Out of the Web Interface You can log out of the cluster from any page in the Violin Web Interface. To log out of your Web Interface session, do the following: 1. From any page in the Web Interface, go to the upper right hand corner. 2. Click the session icon, as shown in the following example.

The Logout page appears acknowledging that you successfully logged out.

3. To log back in, click Login Again and re-enter your credentials.

Accessing and Optimizing the iPad Web Interface The iPad Web Interface is optimized for monitoring the Memory Gateway and cluster nodes. This section demonstrates how to launch the Violin Web Interface on your iPad, and then optimize the display. The graphical user interface (GUI) focuses strictly on monitoring with the following available pages:

• • • •

36

Cluster—For more information, see Manage on page 58. Monitoring—For more information, see Manage LUN Groups on page 76 LUN Monitoring—For more information, see Monitor LUNs on page 110. LUN Status—For more information, see LUN Status on page 51.


535-0008-00 Rev 10

Accessing and Optimizing the iPad Web Interface

Launching the iPad Web Interface This section demonstrates how to launch the Violin Web Interface on an iPad. To launch the Web Interface on your iPad, do the following: 1. Open a Safari browser window. 2. In the browser URL field, enter one of the following:

•

http://

•

http://

•

http://

•

http://

Note: If you specify the IP address or hostname of a cluster node (other than the master), the connection is automatically redirected to the Master Gateway.

The Web Interface log in page appears. 3. Enter a valid username and password in the text fields of the Login page, as shown in the

following example.

4. Click Login. The Web Interface dashboard appears. 5. (Optional) Streamline the dashboard view by following the instructions for Optimizing the iPad

Web Interface on page 37.

Optimizing the iPad Web Interface You can optimize the dashboard view for the iPad and add a launch shortcut icon to your home screen in one easy task. To optimize the iPad Web Interface display and create a shortcut, do the following: 1. Complete the steps as described in Launching the iPad Web Interface on page 37. 2. Click the Add Bookmark icon at the top of the browser window.

535-0008-00 Rev 10


37


3. Choose

Add to Home Screen.

4. Enter a title for the shortcut icon, and click Add. A shortcut icon appears on your home screen. 5. Close the Web Interface session, then click the home screen icon to launch the optimized

version. 6. Continue with Customizing the Dashboard on page 44.

38


535-0008-00 Rev 10

Web Interface Task Maps

Web Interface Task Maps The following tables provide an overview of the Web Interface main menu options, their functions, and the tasks you perform:

• • • •

Home on page 39 Manage on page 40 Monitor on page 41 Administration on page 42

Home Function

Dashboard System View

LUN Status Front and Rear Panel

Tasks

• •

Customizing the Dashboard on page 44

• • •

Viewing Cluster Status on page 49

• •

Viewing Container Status on page 51

• • •

Viewing Rear Panel Information on page 54

Customizing Gadgets on page 47 Viewing VIMM Status on page 49 Viewing VIMMs by RAID Group or Category on page 50 Viewing LUN Status on page 52 Viewing Front Panel Information on page 56 Remotely Activating the ID LED of a Memory Array on page 56

Table 2.5 Home Functions and Tasks

535-0008-00 Rev 10


39


Manage vMOS-6 provides the following Manage features and functions:

• •

Manage LUNs, Initiators, Targets, and LUN Groups available on v5.x and v6.x platforms. vMOS Data Management available only on v6.x platforms.

Manage LUNs, Initiators, Targets, and LUN Groups Function

Manage LUNs

Manage Initiators

Manage Targets Manage LUN Groups

Tasks

•

Searching, Sorting, and Performing LUN Operations on page 60

• • •

Creating and Deleting LUNs on page 62

• • •

Creating and Deleting Initiator Groups on page 69

• •

Viewing Target Status on page 75

• • • •

Viewing LUN Group Information on page 76

Editing LUNs on page 64 Exporting LUNs, Initiator Groups, and Initiators on page 67 Adding or Deleting Initiators from a Group on page 71 Creating, Modifying, and Deleting Nested Initiator Groups on page 72 Managing Transport Protocols on page 75 Creating and Deleting LUN Groups on page 77 Modifying LUN Groups on page 78 Exporting and Unexporting LUNs in a LUN Group on page 79

Table 2.6 Manage LUNs, Initiators, Targets, and LUN Groups

40


535-0008-00 Rev 10

Web Interface Task Maps

vMOS Data Management Function

Thin Provision LUNs

Manage Snapshots

Space Management

Tasks

• • •

Creating and Deleting Thin Provision LUNs on page 81

• • • • •

Working with Snapshot Management on page 84

• • • •

Creating LUN Group Snapshots on page 92

•

Rolling Back a LUN or LUN Group from a Snapshot on page 98

• •

Deleting LUN and LUN Group Snapshots on page 99

• • •

Managing Array Space Usage on page 102

Modifying Thin Provision LUNs on page 83 Exporting and Unexporting Thin Provision LUNs on page 83 Creating Schedules on page 85 Modifying and Deleting Schedules on page 88 Viewing Schedules on page 90 To view a list of exported snapshots, do the following: on page 90 Editing LUN and LUN Group Snapshots on page 95 Exporting LUN and LUN Group Snapshots on page 96 Removing Exports from LUN and LUN Group Snapshots on page 98

Viewing Schedule Calendars on page 100 Managing Container Space Usage on page 103 Managing LUN Space Usage on page 104

Table 2.7 vMOS Data Management

Monitor Function

Monitor Overview Monitor LUNs Monitor LUN Groups Monitor Targets

Tasks

• •

Customizing the Output Display on page 106

• •

Searching for a LUN on page 110

• •

Searching for a LUN Group on page 111

• •

Selecting a Target Port on page 112

Monitoring Container, LUN, and Port Data on page 108 Analyzing Data for a LUN on page 110 Analyzing LUN Group Data on page 111 Analyzing Data for a Target on page 112

Table 2.8 Monitor Functions and Tasks

535-0008-00 Rev 10


41


Administration

Function

Cluster Admin

Tasks

• •

Viewing Software Version and Boot Images on page 114

• • • • •

Changing and Saving Configuration Files on page 115

• •

Managing User Accounts on page 119

• • • • • • •

Viewing Current Network Settings on page 122

• • • • •

Specifying Call Home Settings on page 128

Tools

•

Adding and Removing Feature Licenses on page 139

Information

• • •

Accessing Context-Sensitive Help on page 141

Network

Notification

Rebooting or Shutting Down a Memory Gateway on page 114 Viewing Gateway Node Details on page 115 Viewing Alerts on page 116 Searching for and Viewing Logs on page 117 Generating and Downloading Combined Log Files on page 118 Viewing Node Versions on page 122 Changing the Cluster Management Address on page 123 Modifying Global Default Gateway Settings on page 123 Viewing and Modifying DNS Settings on page 124 Viewing and Modifying NTP Settings on page 125 Viewing and Modifying Web Admin Settings on page 126 Viewing Gateway Node Details on page 115 Managing E-mail Settings on page 129 Managing E-mail Notifications on page 132 Managing SNMP Settings on page 134 Managing SNMP Traps on page 136

Accessing a Full PDF Version of this Guide on page 141 Viewing System, License, and Copyright Information on page 142

Table 2.9 Administration Functions and Tasks

42


535-0008-00 Rev 10

Home Dashboard

Home Dashboard The Home dashboard is a landing page that provides a selection of metrics for tracking system health and performance. The customizable dashboard allows you to easily add, remove, and rearrange the gadgets for at-a-glance system assessment. Likewise, the System View page provides a visual overview of Violin Cluster Status for health and performance. The LUN Status page provides an easy access list of all LUNs and their statistics, while the Front and Rear Panel page

Note: The dashboard supports Internet Explorer 9 (IE 9) and higher. If you are using an earlier

version of Internet Explorer, the dashboard does not display. For more information, see Supported Web Browsers on page 33.

The following table provides a list of tasks and their associated functions. Click a hyperlink to jump directly to the task. Function

Dashboard

System View

LUN Status Front and Rear Panel

Tasks

• • • • • •

Customizing the Dashboard on page 44

• • •

System View on page 48

• •

Viewing Container Status on page 51

• • •

Viewing Rear Panel Information on page 54

Customizing Gadgets on page 47 Viewing Cluster Status on page 49 Viewing VIMM Status on page 49 Viewing VIMMs by RAID Group or Category on page 50 Viewing LUN Status on page 52 Viewing VIMM Status on page 49 Viewing VIMMs by RAID Group or Category on page 50 Viewing LUN Status on page 52 Viewing Front Panel Information on page 56 Remotely Activating the ID LED of a Memory Array on page 56

Dashboard You can select the gadgets that appear on the landing page and arrange them to maximize efficiency for your style of work. You can populate your dashboard with any of the following gadgets.

•

Bandwidth—Shows reads and writes performed in Megabytes per second (MB/s). You can display this information at the container level or LUN level. Click the settings icon in the upper right corner to customize the display.

•

IOPS—Shows read, write, and total statistics in real-time, per hour, or per day values. Click the settings icon in the upper right corner to customize the display. You can display this information at the container level or LUN level.

535-0008-00 Rev 10


43


•

Latency—Shows read, write, and total latency for a container or LUN in real-time, per hour, or per day values. Click the settings icon in the upper right corner to customize the display.

•

Live Stats—Shows real-time values for bandwidth, IOPS and latency on a LUN level or container level. Click the settings icon in the upper right corner to customize the display.

•

Indicators—Shows the LED, temperature and fan status for a container on an array. Click the settings icon in the upper right corner to customize the display.

•

Flash/VIMM Status—Shows the VIMM status (number active, down, etc.) and the remaining length of flash life for a container on an array. Click the settings icon in the upper right corner to customize the display.

•

Container Status—Shows the amount of space used and available on the container, the number of LUNs, and provides an HA status indicator for a container on an array. Click the settings icon in the upper right corner to customize the display.

•

Port Stats—Shows the Received and Sent statistics for InfiniBand and Fibre Channel ports in real-time, per hour, or per day values for a port within a container. Click the settings icon in the upper right corner to customize the display.

•

Snapshots Summary—Shows the number of snapshots, the space used, and the percentage of total space consumed by snapshots. A list of the day’s schedules for snapshots is shown. Click the arrow (>) on the right to jump to the Snapshot Management page and view detailed information on the selected schedule. This gadget is only available with vMOS-6.

Note: For instructions on how to change the information shown on a gadget, see Customizing

Gadgets on page 47.

Customizing the Dashboard You can customize the dashboard by adding, removing, and rearranging the placement of the gadgets. To customize the dashboard and view data, do the following: 1. Log in to the cluster, as described in Logging In and Out of the Web Interface on page 34. The

default dashboard appears. 2. Click the plus icon (+) in the lower right corner allows to access the toolbar. Note: If you are not able to view the dashboard, you may need to upgrade your browser. For more information, see Supported Web Browsers on page 33.

44


535-0008-00 Rev 10

Home Dashboard

The toolbar appears at the bottom of the window, as shown in the following example.

3. Click an icon on the toolbar to add it to the dashboard, then select and drag icons to rearrange.

Click the X icon in the upper right corner of the gadget to remove it from the dashboard display.

535-0008-00 Rev 10


45


4. To display detailed data for a gadget, move your cursor over an area in the graph. A pop-up

window appears, as shown in the following example.

5. To hide the toolbar, click the X icon in the lower right corner of the window.

46


535-0008-00 Rev 10

Home Dashboard

Customizing Gadgets Each dashboard gadgets allows you to customize the display. To customize the information shown on a gadget, do the following: 1. Open the gadget on the dashboard, as described in Customizing the Dashboard on page 44. 2. Click the Settings icon in the upper right corner of the gadget, as shown circled in red in the

following example.

The gadget flips over to show a back panel with options appropriate for that gadget. 3. Make the desired selections and click Save to apply, or Cancel to revert to the previous

settings. The gadget flips back to the front panel display.

535-0008-00 Rev 10


47


System View The System View tab brings up the Violin Cluster Status page, displaying information depicting the health and performance of both Memory Gateways. Easy to read graphics allow you to quickly assess if there is a problem with the system. An expandable Board Status page provides an at-aglance view of VIMM statistics.

Note: A status message appears in the header when a RAID rebuild is in progress

The Violin Cluster Status page provides the following information:

•

Master Memory Gateway Statistics

•

— DRAM and flash allocation — Port ID — Total flash memory — Performance data: read and write rates, DMA active, and pending — Activity graph (reads and writes) for this Memory Gateway — Network data: rates of client and server activity — System drive read and write rates — Additional Memory Gateway statistics: same data as for master Memory Array Statistics — — — — —

48

Number of VIMMs in each state (active, spare, booting, etc.) Activity graph (reads and writes) color-coded for each Memory Gateway Temperatures of ambient air, controller, and VIMMs (for each Memory Gateway) Fan status indicators Estimated lifetime of flash drives Violin 6000 Series Memory Array User’s Guide

535-0008-00 Rev 10

Home Dashboard

•

— Status lights: Alarm, Status, and Ready Network Stats — Client- and server-side statistics. — Measure the throughput of all operations including hits, misses, and write-throughs.

Viewing Cluster Status The System View tab brings up the Violin Cluster Status page. To view cluster statistics, do the following: 1. Select

Home > System View. The Violin Cluster Status page appears.

2. Check alarm status at the right side of the panel. 3. Verify the life span of the flash in the Flash Lifetime panel. 4. Check the health of the Fans. A failed fan displays as red. A problematic fan displays as

yellow. A green box with a white check mark represents a healthy, fully functioning fan. 5. Check the VIMM, Controller, and ambient temperatures. Green is healthy, yellow a caution,

and red a warning. 6. View a high-level synopsis of the flash VIMM statistics in the far left panel.

Viewing VIMM Status You can expand the Violin Cluster Status page to view board-level information about a Memory Array. The Board Status page shows a color-coded array of VIMMs with indicator boxes for alarm status, temperature, and remaining lifetime.You can move your cursor over a VIMM to view a detailed status. To view VIMM statistics, do the following: 1. Select

Home > System View, to display the Violin Cluster Status page.

2. Click the

535-0008-00 Rev 10

Show Board Status tab.


49


3. For general status information about specific VIMMs, move the cursor over the VIMM number. 4. To view Alarm, Temp, or Lifetime information, move the cursor over the appropriate indicator

box.

Viewing VIMMs by RAID Group or Category You can selectively display RAID group information on the Board Status page, as well as category details. To view RAID group and category board-level information, do the following: 1. Select

Home > System View to display the Violin Cluster Status page.

2. Click the


3. To view VIMMs by RAID group, select a RAID Group number at the top-center of the Board

Status panel. Selecting a RAID Group highlights the VIMMs belonging to that RAID Group in the Board Status panel, as shown in the following example.

50


535-0008-00 Rev 10

Home Dashboard

4. To view failed VIMMs, select the

Failed link in the Show area of the Board Status panel.

A failed VIMM appears orange, as shown in the following example.

5. To view booting VIMMs, select the

Booting link in the Show area of the Board Status panel.

6. To view admin down VIMMs, select the Admin Down link in the Show area of the Board Status

panel. 7. To view spare VIMMs, select the

Spares link in the Show area of the Board Status panel.

LUN Status The LUN Status page allows you to view information about the LUNs within a container. Every LUN is created within a storage container (array).

Viewing Container Status The LUN Status page provides easy access to information about the container in an array. At-aglance information includes the container name, its address, and the number of LUNs within the container. The following task demonstrates how to view more detailed information. To view LUN and container status, do the following: 1. Select

Home > LUN Status. The LUN Status page appears.

2. Select the

535-0008-00 Rev 10

Container icon to view real-time statistics about the container.


51


3. Select the LUN icon to view real-time statistics about the LUN. 4. Click the container name to view detailed information about the container and the LUNs it

contains. The LUN Management page appears, with the container information in the upper left panel.

Note: For more information on managing LUNs, see "Manage LUNs" on page 59.

Viewing LUN Status The LUN Status page provides easy access to information about the container in an array. At-aglance information includes the LUN name, size, and the number of active sessions. The following task demonstrates how to view more detailed information. To view LUN and container status, do the following: 1. Select

Home > LUN Status. The LUN Status page appears.

2. Select a

52

LUN icon to view real-time statistics about the container.


535-0008-00 Rev 10

Home Dashboard

3. Select the LUN icon next to the name to view real-time statistics about that LUN. 4. Click the LUN name link to view detailed information about that LUN.

The LUN Management page appears, with the selected LUN highlighted in the LUN panel on the left. In the top right panel, details of the selected LUN are shown.

Note: For more information on managing LUNs, see Manage LUNs on page 59.

535-0008-00 Rev 10


53


Front and Rear Panel The Front and Rear Panel page provides a real-time view of the active ports (shown in green) that are currently used by the system (Rear Panel View) and the LEDs on the front of the Memory Array (Front Panel View). The port-specific information is updated automatically every 30 seconds. You can also remotely activate the ID LED from both the front- and rear-panel views, which is helpful when you need to identify a particular Memory Array in an equipment rack. Viewing Rear Panel Information You can view information about each active port, from the port name to specific information about the port, including details about the Memory Gateways, NIC ports, Ethernet ports and power supplies. To view the rear panel of the array, do the following: 1. Select

Home > Front and Rear Panel.

The Front and Rear Panel page displays, with an image of the back of the Memory Array shown at the top.

2. Rest the mouse pointer over a port (power supply, Ethernet connector, Network Interface port)

on the rear panel image to identify the connector. A ToolTip shows the name and type of port selected. 3. Click the image to show more detailed information and real-time statistics. Note: The PCIe ports are used for attaching the Memory Array directly to a host machine without

going through the Memory Gateways (NIC ports). Since the Web interface is run from the Memory Gateways, PCIe port information is unavailable in the Web interface.

54


535-0008-00 Rev 10

Home Dashboard

The active port area selected is displayed in the lower left panel of the page. The following example shows the lower left and right panels for an InfiniBand configuration.

4. If specific port information is available, click a port in the lower left panel to view the

information in detail. Specific information is available for the following ports:

•

Power Supplies A and B: The power supply’s status, and current voltage and amps are shown.

•

Ethernet Ports A and B: View the Memory Gateway Management IP address; host names and IP addresses of the Master and Standby Gateways; the ACM Management IP address; host names and IP addresses of the Master and Standby ACMs.

•

Network Interface Ports: View the Memory Gateway that manages the port (mg-a or mg-b) and the port’s state and speed. Also available is an Port Stats gadget, which shows the Received and Sent statistics for InfiniBand and Fibre Channel ports in real-time, per hour, or per day values for a port within a container. If you select the active Network Interface Ports area on the Rear Panel image, the lower left panel that displays shows the slot and port numbers for the cards installed in the system, along with the representative CLI values for each port.

Note: The physical port numbers on the network interface cards (NICs) are not aligned with their corresponding port names in the CLI. For example, in a Fibre Channel environment, NIC port A1 is shown as “hba-a1” in the CLI, and NIC port C1 is also shown as “hba-a1” in the CLI. This is because the two Memory Gateways, shown with their CLI representations in the lower left panel (mg-a and mg-b), separately manage two NICs each. To verify the actual port numbering shown in the CLI, log in to the Memory Gateway master and then run the “show targets” command.

To switch the view to the front panel, Click the Front Panel button in the upper right of the Rear Panel View.

535-0008-00 Rev 10


55


Viewing Front Panel Information The Front Panel view shows the front of the Memory Array and its USB ports and LEDs. The front panel contains two USB ports (A and B), the ID button, and six LEDs (AC-A, AC-B, System Status, VIMM Status, Port/Interface Status, ID). To view the front panel of the array, do the following: 1. Select

Home > Front and Rear Panel.

The Front and Rear Panel page displays, with an image of the back of the Memory Array shown at the top. 2. Click the Front Panel button in the upper right of the Rear Panel View image to switch the

view to the front panel. The front panel image displays, with the ports and LEDs highlighted on the front panel image. A zoomed image displays below the front panel.

3. Rest the mouse pointer over a USB port or LED to view the name of the port or LED. The

following ports and LEDs are viewable:

•

USB ports: The USB ports are used for performing software upgrades.

•

AC-A and AC-B LEDs: The Master ACM (AC-A or AC-B) is shown in green. Use its USB port when upgrading the software from a USB drive.

•

SYS (System Status) LED: This LED is green when all system modules are operating normally, amber when an alarm has been raised, and grey when the Memory Array is off or starting up.

•

VIMM LED: This LED is green when all VIMMs are functioning correctly, and amber when an alarm has been raised for at least one VIMM, or when a VIMM is not powered or present.

•

ID LED/Button: This LED is blue when the button has been pressed or when it has been activated remotely using the Violin Web interface. See Remotely Activating the ID LED of a Memory Array on page 56.

To switch the view to the rear panel, Click the Rear Panel button in the upper right of the Front Panel View. Remotely Activating the ID LED of a Memory Array The ID LED can be remotely activated from both the front- and rear-panel views. The ID LED is used to locate a particular Memory Array in an equipment rack, which is helpful when a Memory Array needs attention and the equipment rack holds more than one Array.

56


535-0008-00 Rev 10

Home Dashboard

To activate the ID LED from the Rear Panel View, do the following: 1. Select

Home > Front and Rear Panel. The Front and Rear Panel page appears.

2. Click the lower left of the Rear Panel image to remotely activate the ID LED of this Memory

Array.

The LED is activated when the LED is blue on the rear panel image. 3. Click the blue LED to turn off the ID LED.

To activate the ID LED from the Front Panel View, do the following: 1. Select

Home > Front and Rear Panel. The Front and Rear Panel page appears.

2. Click the Front Panel button in the upper right of the Rear Panel View image to switch the

view to the front panel. 3. Click the ID button to remotely activate the ID LED of this Memory Array.

The LED is activated when the LED is blue on the front panel image. 4. Click the blue ID button to turn off the ID LED.

535-0008-00 Rev 10


57


Manage You can manage LUNs, initiators, targets, and LUN groups, as well as utilizing data management functionality. The following tables provide a list of tasks and their associated functions. Click a hyperlink to jump directly to the task. vMOS-6 provides the following Manage features and functions:

• •

Manage LUNs, Initiators, Targets, and LUN Groups available on v5.x and v6.x platforms. vMOS Data Management available only on v6.x platforms.

Manage LUNs, Initiators, Targets, and LUN Groups Function

Manage LUNs

Manage Initiators

Manage Targets Manage LUN Groups

Tasks

•

Searching, Sorting, and Performing LUN Operations on page 60

• • •

Creating and Deleting LUNs on page 62

• • •

Creating and Deleting Initiator Groups on page 69

• •

Viewing Target Status on page 75

• • • •

Viewing LUN Group Information on page 76

Editing LUNs on page 64 Exporting LUNs, Initiator Groups, and Initiators on page 67 Adding or Deleting Initiators from a Group on page 71 Creating, Modifying, and Deleting Nested Initiator Groups on page 72 Managing Transport Protocols on page 75 Creating and Deleting LUN Groups on page 77 Modifying LUN Groups on page 78 Exporting and Unexporting LUNs in a LUN Group on page 79

Table 2.10 Manage Functions and Tasks

58


535-0008-00 Rev 10

Manage

vMOS Data Management Function

Thin Provision LUNs

Manage Snapshots

Space Management

Tasks

• • •


• • • • •


• • • •


•


• •


• • •




Table 2.11 Data Management Functions and Tasks

Manage LUNs The following panels comprise the LUN Management Page:

•

Container—Displays the container name, the percentage of space used, the number of LUNs in the container, Gateway A and B names, HA status, active online ports, and a bar graph depicting container space usage.

•

LUNs—Shows information about the LUNs within the selected container, including name, serial number (exported LUNs only), provisioned size, allocated size, and total number of snapshots for the LUN, if any. You can create LUNs, search for a LUN, reorder the list, and perform actions on LUNs by selecting one or more check boxes. When you highlight a LUN (selecting it so that it changes color), detailed information about the LUN appears in a panel on the upper right. Note: Only one LUN can be highlighted at a time.

•

Highlighted LUN—Shows detailed information on the highlighted LUN (in the LUNs panel), including if it is read-only, and/or has Port A or Port B offline. You can view and add or remove Exports for a selected LUN.

535-0008-00 Rev 10


59


Caution: You can change the online status of Port A and Port B (Gateway A and Gateway B) in

the highlighted LUN information panel. However, this is generally not recommended.

Searching, Sorting, and Performing LUN Operations A search (filter) operation is performed across all columns simultaneously to return the desired results. Sorting applies to the selected column. However, the sorting changes the order of all the records (rows/columns).

Note: You can select multiple LUNs and perform an operation on them simultaneously. A check

mark in the box next to the LUN name shows that a LUN is selected.

To search for a LUN and customize the display, do the following: 1. Select

Manage > Manage LUNs. The LUN Management page appears.

2. To search for a LUN, enter a string in the Search field. A list of LUNs that contain the string

appear in the LUNs panel.

3. Select a LUN to highlight it and view details in the upper right. 4. Click the down arrow in the header to view the list in descending order, by name (as shown in

the following example), serial number (for exported LUNs), provisioned size, allocated size, or number of snapshots. Click the up arrow in any of the columns to view the list in ascending order (the default display).

60


535-0008-00 Rev 10

Manage

5. To change the number of entries shown, choose one of the following from the Show entries

drop-down list: 10, 25, 50, or 100. The default is 10. To perform operations on one or more LUNs, do the following: 1. Select


2. In the LUNs panel, select the check boxes next to the desired LUNs. A check mark appears in

the box next to the name. Note: A highlighted LUN that is not selected is not included in an operation. Performing an operation on multiple LUNs simultaneously may take some time. 3. Select one of the following options to apply the operation to all checked LUNs:

•

Remove Checked LUN(s)

• •

Remove Exports for Checked LUN(s) Export Checked LUN(s)

In the following example, Export Checked LUNs is selected. For more information on how to export LUNs, see Exporting LUNs, Initiator Groups, and Initiators on page 67.

The highlighted LUN in this example does not show a check mark in its check box. Therefore, it is not included in the operation.

535-0008-00 Rev 10


61


Creating and Deleting LUNs This section walks you through the process of creating a LUN. Every LUN is created within a container. The size and amount of used space in the container may restrict the number and size of the LUNs created within the container. Prerequisites

Verify the block size that is acceptable for the clients (initiators) on your network. Disk storage systems previously assumed a 512-byte sector or block size as the unit to store data. Newer devices, such as Violin Memory Arrays, have a native storage unit sized at 4096 bytes but can also emulate 512 bytes. Not all client operating systems are able to understand accessing anything other than a 512-byte based disk drive. To be safe, choose a 512 block size if you are not sure what type of clients will be connected to an exported LUN. For example, RHEL5 and Windows Server 2003 are not be able to properly use a 4096 block-sized disk, and this selection leads to unexpected behavior. For newer operating systems which support a 4096 block size, the emulation layer can be avoided to provide optimal performance.

Note: VMware ESX 4.1 does not support 4K block size. For VMware ESX 4.1, use a 512 byte

block size LUN, which is supported.

To create a LUN, do the following: 1. Select


2. In the LUNs panel, click the

Create New LUNs icon.

The Create LUN window appears. 3. Specify the following:

• •

The number of LUNs in the # LUNs to Make field. The name of the LUNs in the LUN Name field.

Note: When creating more than one LUN, the name of each new LUN is appended with a number

beginning with 1, and continuing through the number of specified LUNs. This ensures that each LUN has a unique name. Creating a large number of LUNs at once (50 LUNs or more), the operation may take some time.

62


535-0008-00 Rev 10

Manage

4. Define the size of a LUN, in one of the following ways:

•

To define the size of the LUN based on the space available, select Use all available

space equally. •

To manually define the size of each LUN, select Specific size per LUN and enter the size in GB.

Note: The size of the LUNs may be restricted by the space available in the container. 5. Select one of the following block sizes for the LUN:

•

512 bytes.

•

4096 bytes.

Caution: Not all systems can handle 4096-byte blocks. Use 512-byte blocks, if you are not sure what type of clients will be connected to an exported LUN. 6. Specify the following settings, as appropriate:

•

Read-only—Sets the LUNs as read-only.

•

Thin Provisioning—Creates a thin LUN. For more information, see Thin Provisioning on page 260.

•

NACA—Provides a performance enhancement for IBM AIX systems. For more information, see AIX Compatibility NACA Bit Support on page 179.

7. Click

OK. The specified LUNs are created in the container.

To delete a LUN, do the following: 1. Select


2. Search for the LUN, as described in Searching, Sorting, and Performing LUN Operations on

page 60.

535-0008-00 Rev 10


63


3. Select the LUN so that it highlights, then remove the LUN exports in the following way: a. In the Exports panel on the right, click the check boxes for each of the exports associated

with the LUN. b. Click the

Unexport icon. Then click OK in the pop-up dialog that appears to confirm the

action. Note: All exports must be removed before the LUN can be deleted.

4. Click the check box of the LUN you want to delete. A check mark appears. Note: The check box must display a check mark for the LUN to be deleted. Simply highlighting the selection does not include it in the operation. 5. In the LUNs panel, click the Delete icon. Then click

OK in the pop-up dialog that appears to

confirm the action.

Editing LUNs You can change the name of an existing LUN and increase the size of the logical volume. Decreasing the size of an existing LUN is not supported. For a smaller size LUN, create a new LUN of the desired size. For more information, see Creating and Deleting LUNs on page 62.

WARNING! Existing file systems on the client side are not extended as part of the resize operation. File systems must be extended from the client initiator. Not all client systems will detect the LUN size change until they reboot.

To modify access to an existing LUN, do the following: 1. Select


2. Select a LUN so that it highlights. Details appear in the top-right LUN information panel.

64


535-0008-00 Rev 10

Manage

3. In the LUN information panel make the following selections, as necessary:

— ALUA—Enables or disables Asymmetric Logical Unit Access. For more information, see Using Asymmetric Logical Unit Access (ALUA) on page 66. — Read Only—Activating read-only access on an existing LUN can cause data loss for connected clients with cached data. — Online Ports—Unchecking these options takes the LUN offline for the respective Memory Gateway nodes. — NACA—provides a performance enhancement for IBM AIX systems. For more information, see AIX Compatibility NACA Bit Support on page 179. Caution: After the NACA option is enabled, you should only export the LUN to an AIX host.

To change the name of an existing LUN, do the following: 1. Select


2. Select the LUN so that it highlights. Note: All exports should be removed before changing the name of the LUN. 3. Remove any LUN exports in the following way: a. In the Exports panel on the right, click the check boxes for each of the exports associated

with the LUN. b. Click the Delete icon. Then click OK in the pop-up dialog that appears to confirm the action. 4. Click the

Edit LUN icon.

5. In the Edit LUN dialog, click Rename LUN.

535-0008-00 Rev 10


65


6. Enter the new name in the

New LUN Name text box, and click OK.

The new name is applied and appears in the LUNs panel. To increase the size of an existing LUN, do the following: 1. Select


2. Select the LUN so that it highlights, then click the 3. In the Edit LUN dialog, click

Edit LUN icon.

Resize LUN.

Note: You can increase the size of a LUN. However, decreasing the size is not allowed. 4. Enter the

5. Click

New LUN size (GB) in the text box.

OK.

The LUN size is increased and appears in the LUNs panel.

Using Asymmetric Logical Unit Access (ALUA) Asymmetric Logical Unit Access (ALUA) is a multipathing technology that enhances availability through efficient data access. When enabled, ALUA communicates to clients when certain paths to data become unavailable. This allows the client host to switch to available paths and perform necessary actions efficiently. For information on using ALUA for data management, see Optimizing Performance with ALUA on page 257.

66


535-0008-00 Rev 10

Manage

Note: ALUA is not enabled by default. For optimum performance, it is recommended that you

enable ALUA on thin provisioned LUNs and LUNs with snapshots.

To enable ALUA, do the following: 1. Select


2. In the LUNs panel, highlight the LUN for which you want to enable ALUA. Information about

the highlighted LUN appears in the top-right panel. Note: You do not need to select the LUN check box. 3. In the LUN information panel, click the ALUA check box.

4. Click OK in the pop-up dialog that appears, to confirm the action. 5. Repeat the process to enable ALUA on other LUNs in the container.

Exporting LUNs, Initiator Groups, and Initiators This section demonstrates how to export LUNs, initiator groups, and initiators. Be aware of the following guidelines for exporting LUNs. For information on nested initiator groups, see Creating, Modifying, and Deleting Nested Initiator Groups on page 72 and Managing Nested Initiator Groups on page 213.

•

Changing the LUN ID after a LUN has been exported is not recommended, as traffic is disrupted during the change. The effect on the client depends on how it handles LUN ID changes.

•

You can export LUNs to different igroups on different targets, even when they have initiators in common. This is an enhancement for vMOS-6.0 only.

Note: After you create an export for a specific iSCSI initiator using a specific target, the initiator

is excluded from seeing LUN exports from that target to all initiators.

To export LUNs, do the following: 1. Select


2. In the LUNs panel, select the check box next to the LUN(s) you want to export.

535-0008-00 Rev 10


67


Note: The check box next to the LUN name must show a check mark for the LUN to be included in the export. 3. Click Export Checked LUNs. The Add Export dialog box appears.

4. Do one of the following to select the initiators to which the LUNs will be exported.

•

To export to all initiators, select the All Initiators option button.

•

To export to select initiator groups or initiators, select Specific Initiator Groups and/or Initiators. Select one or more Initiator Groups from the list, and enter and select initiator names (each on their own line) in the Individual Initiator panel.

5. Specify the target ports through which the LUNs are exported in one of the following ways:

•

Select All Ports, to export through all target ports.

•

Select Specific Ports, to export through target ports selected from the list.

6. Assign IDs to the exported LUNs in one of the following ways:

•

Select Value and enter a value in the text field to manually assign a LUN ID. User-defined LUN IDs can make it easier to differentiate between LUNs. If you do assign a user-defined LUN ID, Violin Memory recommends that you specify a number below 255 as some operating systems (for example, Windows) only discover LUN IDs between 0 and 254. This section is automatically disabled when you create multiple LUNs at one time.

•

68

Select Auto to automatically assign an ID to the exported LUNs. This option is mandatory when you export multiple LUNs at once.


535-0008-00 Rev 10

Manage

Caution: Re-exporting a LUN with an auto-assigned LUN ID may cause the system to assign a new LUN ID to the export, which may disrupt the existing I/O traffic. The same holds true for manual LUN ID changes. 7. Click

OK. Initiator and target information for the exported LUNs appears in the Exports panel on the right side of the window.

Manage Initiators In a Violin Memory environment, hosts (such as database servers or application servers) that access LUNs are initiators and the arrays are the targets. You can control initiator access to target LUNs in the following ways:

• • •

Specifying access on an individual initiator basis Specifying access through a group of associated initiators, known as an igroup Specifying access through a group of igroups, known as nested igroups

The use of igroups streamlines initiator management by allowing you to define a group of initiators (igroup), and then assign that group access to select LUNs. So instead of specifying access for each individual initiator, you specify access for the igroup and all the initiators that are members of the igroup are granted access to the LUNs. The Web Interface allows you to easily create and modify initiator groups, as described in Creating and Deleting Initiator Groups on page 69 and Adding or Deleting Initiators from a Group on page 71. You can also combine initiator groups within another initiator group, as described in Creating, Modifying, and Deleting Nested Initiator Groups on page 72.

Note: Check with your network administrator before adding initiator hosts in the Advanced panel,

and verify the correct addresses of the hosts.

Creating and Deleting Initiator Groups You can create and delete Initiator groups from the Manage > Manage Initiators page of the Web Interface. A list of initiator groups is shown in the left panel, with a list of available sessions shown in the right panel for the highlighted initiator group. You can also add initiator hosts, as described in Adding or Deleting Initiators from a Group on page 71. The All igroups and Initiators panel provides a list of available initiators and igroups, along with the following information for each: — — — — —

535-0008-00 Rev 10

Name: Name of the initiator or igroup. Type: Initiator or igroup Export: LUNs exported to the igroup or initiator, or not igroups: Number of igroups in an igroup Initiators: Number of initiators that belong to the igroup


69


To create an initiator group, do the following: 1. Select 2. Click

Manage > Manage Initiators. The Manage Initiators and igroups page appears.

Create igroup.

The Create New igroup dialog appears. 3. Enter a name in the igroup name text box, then select the check boxes for the initiators and

igroups to include in the new igroup, and click Create. Note: To add initiators to your new igroup that do not appear in the list, see Adding or Deleting Initiators from a Group on page 71.

Note: Adding an existing igroup to a new igroup creates a nested igroup. For more information,

see Creating, Modifying, and Deleting Nested Initiator Groups on page 72.

To delete an initiator group, complete the following steps: 1. Select

Manage > Manage Initiators.

2. Select the check box for one or more initiator groups you want to delete.

A check mark appears next to the name of a selected igroup. Only igroups with a check mark next to the name will be deleted.

70


535-0008-00 Rev 10

Manage

3. Click

Delete.

A dialog appears prompting you to confirm the action. 4. Click

OK.

A prompt appears verifying the success of the operation, and the selected initiator groups are removed from the list.

Adding or Deleting Initiators from a Group You can select initiator hosts to add to an existing initiator group from the panel on the right. Or from the Advanced panel, you can add initiator types that do not appear in the list (iSCSI), or that may not yet be on the network. This section demonstrates how to add initiator hosts to an igroup by selecting them from the list, and then add initiators using the Advanced panel.

Caution: Adding an initiator to multiple igroups can create conflicts, and cause problems in

export behavior.

Prerequisites

When adding initiators, verify the validity and availability of the address. Check with your system administrator regarding the availability status of the initiators. Each initiator address has a protocol-specific identifier:

•

iSCSI initiatory identifier formats: iqn (most common), naa, eui For example: iqn.1994-05.com.redhat:xyz

•

Fibre Channel initiator identifier format: wwn Fibre Channel initiators are fixed by the appropriate HBA port, for example: wwn.01:02:03:04:05:06:07:08

•

InfiniBand initiator identifier format: guid InfiniBand initiators are fixed by the appropriate HCA port, for example: guid.0123:0456:0789:123

To add or remove initiators from an existing igroup, do the following: 1. Select

535-0008-00 Rev 10

Manage > Manage Initiators. Violin 6000 Series Memory Array User’s Guide

71


2. Highlight the initiator group in list that you want to modify. The initiators and igroups that are

members of that group appear in the panel on the left. 3. Click

Edit igroup. The Edit igroup dialog appears.

4. To remove an initiator host or igroup from the group, deselect a check box. 5. To add initiator hosts that are not on the list, under Custom Initiators, do the following: a. Click

More.

b. In the text box at the bottom of the window, enter a valid initiator host address. c. Press

Enter to add another address. Each address must be on its own line.

6. To apply your changes, click

Modify.

Caution: Custom initiator hosts appear in the group even if the host has not yet been added to

the network. Check with your network administrator for status on the availability of these hosts.

Creating, Modifying, and Deleting Nested Initiator Groups A nested initiator group (nested igroup) is an igroup that contains one or more igroups or initiators. This concept is similar to a folder that contains other folders, or subfolders. For more information on nested initiator groups, see Managing Nested Initiator Groups on page 213.

Note: You can create up to three hierarchy levels within a nested igroup.

72


535-0008-00 Rev 10

Manage

Prerequisite

You must have created igroups, as described in Creating and Deleting Initiator Groups on page 69. To create a nested igroup, do the following: 1. Select

Manage > Manage Initiators. The Manage igroups and Initiators page appears.

2. In the All igroups and Initiators panel, highlight the igroup in which you will nest other igroups,

then click Edit igroup. The Edit igroup window appears. 3. Select a check box for an igroup in the list to add it to the igroup. If the igroup does not appear

in the list of items on the first page: — Search for the igroup by entering a text string in the Search text box, and pressing Enter. — Click the right-facing arrow at the bottom right corner of the page, and repeat to page through all items on the list. 4. Click

Modify.

5. A prompt appears to tell you that the operation was successful. Click

OK.

The results of the modification appear in the hierarchy tree on the right, as well as in the igroups and Initiators columns to the left.

535-0008-00 Rev 10


73


To modify a nested igroup, do the following: 1. Select


2. In the All igroups and Initiators panel, highlight the nested igroup you want to modify, then click

Edit igroup. 3. Select a check marked box to remove the item from the igroup. Select other check boxes to

add those items to the nest. 4. When you are satisfied with the changes, click

Modify.

5. Click

OK to verify the success of your actions. The results appear in the Manage igroups and Initiators panel.

To deconstruct a nested igroup, do the following: 1. Select


2. In the All igroups and Initiators panel, highlight the nested igroup you want to deconstruct (i.e.

remove all nested igroups) then click Edit igroup. 3. Select the check marked box, or boxes, of the nested igroups you want removed, then click

Modify. 4. Click

OK to verify the success of your actions. The results appear in the Manage igroups and Initiators panel.

To delete a nested igroup, do the following: Note: If any LUNs are exported to the nested igroup, the deletion is not permitted. There will not

be a check box next to the nested igroup in this case. 1. Select


2. In the All igroups and Initiators panel, select the check box for the nested igroup you want

removed. Note: Be sure to deselect the check marked boxes of items you do not want removed. 3. Click

74

Delete. The nested igroup is removed, and disappears from the list.


535-0008-00 Rev 10

Manage

Manage Targets In a cluster, each Memory Gateway operates as a SAN (Fibre Channel/InfiniBand), or LAN (iSCSI) target that provides access to the LUNs stored on its attached memory arrays. Fibre Channel, InfiniBand, and iSCSI target ports are all are supported. Every target is either a specific port on a hardware Fibre Channel host bus adapter (HBA), InfiniBand host channel adapter (HCA), or a software-configured target for iSCSI.

Viewing Target Status On the Target Management page, the details for the highlighted port appear in the panel on the right. The node, target name, hostname, speed, and address of each type of target port. To view status information for a target, do the following: 1. Select

Manage > Manage Targets. The Target Management page appears, showing information for all available target ports in one table.

2. To view details on a specific target, select (highlight) the target in the Available Targets panel

on the left.

Managing Transport Protocols Some target port types are configured automatically when a storage container is created. Other target ports must be configured manually.

•

Fibre Channel target ports are automatically configured when you create the storage containers on Violin Memory Arrays.

•

InfiniBand target ports are automatically configured when you create the storage containers on the Violin Memory Arrays.

•

iSCSI target ports must be configured manually.

To enable target ports, do the following: 1. Select

Manage > Manage Targets. The Target Management page appears.

2. Click the appropriate Enable protocol button.

To a disable active target ports, do the following: 1. Select

535-0008-00 Rev 10

Manage > Manage Targets. The Target Management page appears.


75


2. In the Manage Target Protocols panel, click the appropriate Disable protocol button.

Manage LUN Groups You can create a LUN group with one or more LUNs. When you perform an action on one LUN in a group, that action is applied to all LUNs that are members of the group. A LUN group typically consists of LUNs that are used by a single application. For example, a database application may require two or more LUNs for data and logs. This section covers the following topics:

• • • •

Viewing LUN Group Information on page 76 Creating and Deleting LUN Groups on page 77 Modifying LUN Groups on page 78 Exporting and Unexporting LUNs in a LUN Group on page 79

For information on LUN group performance monitoring, see Monitor LUN Groups on page 111.

Viewing LUN Group Information This section demonstrates how to view LUN and LUN group information using the Web interface. To view LUN group information, do the following: 1. Select

Manage > Manage LUN Groups.

2. To view LUN group information, select a name in the LUN Group List so that it changes color

(highlights). A list of LUNs belonging to the group appears in the right panel.

3. (Optional) Change the order of the list (ascending/descending) by clicking the Up or Down

arrow in the column heading.

76


535-0008-00 Rev 10

Manage

Creating and Deleting LUN Groups Keep in mind the following guidelines when creating LUN groups:

• • •

A LUN may be part of multiple LUN groups.

•

A LUN group name must be unique within a container.

A LUN group can contain a minimum of one LUN and a maximum of eight LUNs. A LUN group name must start with an alphanumeric character, and may contain dashes, underscores, and periods.

Note: Deleting a LUN group does not effect the LUNs belonging to the group, it only deletes the

structure that associated the LUNs as a group entity.

To create a LUN group, do the following: 1. Select

Manage > Manage LUN Groups. The LUN Group Management page appears.

2. Click the

Create New LUN Group icon.

3. In the Create a new LUN Group dialog, enter a unique LUN Group Name in the text field, then

add a description in the Description text field. 4. (Optional) To search for a LUN to add or delete, enter a text string in the appropriate search

box. The results appear in the respective list. 5. To add a LUN to the group, click the Add arrow to the right of the LUN name in the Available

LUNs list. Likewise, to remove a LUN from the group, click the Remove radio button in the Selected LUNs list.

6. Click OK to create the LUN group. The new LUN group appears in the LUN Group List.

535-0008-00 Rev 10


77


To delete a LUN group, do the following: 1. Select

Manage > Manage LUN Groups. The LUN Group Management page appears.

2. In the LUN Group List, select the LUN group so that it highlights, changing color. 3. Click the

Delete LUN Group icon.

Note: Deleting a LUN group does not effect the LUNs belonging to the group, it only deletes the

structure that associated the LUNs as a group entity.

4. In the pop-up dialog, click

Yes to delete the selected LUN group, or No to cancel the

procedure.

Modifying LUN Groups This section demonstrates how to modify an existing LUN group, by adding and removing LUNs from the group.

Note: A LUN group can contain a minimum of one LUN and a maximum of eight LUNs.

To modify a LUN group, do the following: 1. Select

Manage > Manage LUN Groups.

2. In the LUN Group List, select the LUN group you want to modify so that it highlights, changing

color. 3. Click the

78

Edit LUN Group icon.


535-0008-00 Rev 10

Manage

4. (Optional) To search for a LUN to add or delete, enter a text string in the appropriate search

box. The results appear in the respective list. 5. In the Edit LUN Group dialog, do the following as necessary:

— Click the Add arrow in the Available LUNs list to add the associated LUN to the group. — Click the Remove radio button in the Selected LUNs list to delete the associated LUN from the group.

6. Click OK to accept the changes, or Cancel to abort the operation without making any changes.

Exporting and Unexporting LUNs in a LUN Group Exporting and unexporting LUNs that belong to a LUN group follows the same guidelines and procedures as those or LUNs that do not belong to a LUN group. These procedures are accomplished from the Manage > Manage LUNs page. For more information, see Exporting LUNs, Initiator Groups, and Initiators on page 67. Likewise, the procedure for unexporting LUNs that belong to a LUN group is the same as that used for regular LUNs. Likewise, this procedure is accomplished from the Manage > Manage LUNs page. For more information, see Exporting and Unexporting Thin Provision LUNs on page 83.

Data Management vMOS Data Management allows you to manage and protect sensitive data, as well as simplify backup and disaster recover with the following features.

• • • •

Thin provision LUNs for better utilization of storage Snapshots of individual LUNs and LUN groups, read-only or read-write Schedules for automated snapshots Manage and monitor provisioned and allocated space on the array

For a list of data management terminology and use case scenarios, see CHAPTER 8, “Data Management” on page 253.

535-0008-00 Rev 10


79


The following table provide a list of tasks and their associated functions. Click a hyperlink to jump directly to the task. Function

Thin Provision LUNs

Manage Snapshots

Space Management

Tasks

• • •


• • • • •


• • • •


•


• •


• • •




Thin Provision LUNs Thin provisioning provides on-demand allocation of blocks of data, versus the traditional method of allocating all the blocks up front. For example, you can specify the (provisioned) size of a thin LUN, but space is not physically allocated until the data is written to the LUN. For more information, see CHAPTER 8, “Data Management” on page 253.

80


535-0008-00 Rev 10

Manage

Thin LUNs and thick LUNs can be easily distinguished by their icons. Thick LUNs have a solid icon to the left of the name, while thin LUNs have a partially transparent icon, as shown in the following example.

Thick LUN Thin LUN Figure 2.2 Thick LUN and Thin LUN Icons.

This section covers the following topics:

• • •

Creating and Deleting Thin Provision LUNs on page 81 Modifying Thin Provision LUNs on page 83 Exporting and Unexporting Thin Provision LUNs on page 83

Creating and Deleting Thin Provision LUNs You can specify the provisioned size of a thin LUN, but the space is allocated as the data is written. This allows you to use space as needed.

Note: A LUN can be created as thin or thick. However, once a LUN is created as a certain type (thick or thin), the type cannot be changed. A thick LUN remains thick and a thin LUN remains thin until deleted.

To create a thin provisioned LUN, do the following: 1. Choose


2. On the LUNs menu bar, click the

Create New LUN icon.

The Create LUNs dialog appears. 3. Specify the following information:

•

Number of LUNs to Make. The default is 1

•

A unique LUN name

535-0008-00 Rev 10


81


•

The amount of allocated space (in Gigabytes) for each LUN, or use all available space equally across the number of LUNs you are creating

•

512 byte block size, or 4096 byte block size

4. Select the

5. Click

Thin Provisioning check box, along with other appropriate Settings options.

OK. The newly created thin provision LUN appears at the top of the LUN list with a thin

LUN icon to the left of the name.

Note: An exported LUN cannot be deleted. You must unexport the LUN first.

To delete a thin provision LUN, do the following: 1. In the LUNs list, select the check box of the LUN you want to delete. A check mark appears in

the check box. Note: A check mark must appear in the check box beside a LUN for it to be deleted. Highlighting the LUN so that it changes color is not sufficient. 2. In the LUNs menu bar, click the trash can icon.

82


535-0008-00 Rev 10

Manage

3. In the Remove LUN Warning dialog, click

OK to delete the selected LUN, or Cancel to abort

the procedure.

If you confirm the deletion, the thin LUN disappears from the LUN list.

Modifying Thin Provision LUNs You can modify an existing thin provision LUN in the following ways:

• •

change the LUN name increase the allocated space for the LUN

You perform these procedures the same way you would for a thick LUN. For more information, see Editing LUNs on page 64.

Caution: Changing the name of an exported LUN (thin or thick) is not recommended.

Decreasing the size of an existing thin provision LUN is not supported.

Exporting and Unexporting Thin Provision LUNs The procedure for exporting thin provision LUNs is the same as that used for regular (thick) LUNs. For more information, see Exporting LUNs, Initiator Groups, and Initiators on page 67. Likewise, the procedure for unexporting thin provision LUNs is the same as that used for regular LUNs. For more information, see Exporting and Unexporting Thin Provision LUNs on page 83.

Manage Snapshots The Snapshot Management page allows you to view and manage snapshots for LUNs and LUN groups. You can also view and manage the schedules used to create snapshots and snap groups. This section covers the following topics:

• • • • • • • •

Snapshot Rules on page 84 Working with Snapshot Management on page 84 Creating Schedules on page 85 Modifying and Deleting Schedules on page 88 Viewing Schedules on page 90 To view a list of exported snapshots, do the following: on page 90 Creating LUN Group Snapshots on page 92 Editing LUN and LUN Group Snapshots on page 95

535-0008-00 Rev 10


83


• • • • •

Exporting LUN and LUN Group Snapshots on page 96 Removing Exports from LUN and LUN Group Snapshots on page 98 Rolling Back a LUN or LUN Group from a Snapshot on page 98 Deleting LUN and LUN Group Snapshots on page 99 Viewing Schedule Calendars on page 100

Snapshot Rules The following rules apply for snapshots:

• • • •

A LUN can have a maximum of 1,024 snapshots

•

Deleting a single snapshot from a snap group is not allowed

A container can have a maximum of 10,000 snapshots Taking a snapshot of a snapshot is not allowed Snapshots are taken one at a time (serial); creating multiple simultaneous snapshots is not allowed at this time

Caution: Thin provisioned LUNs and LUNs with snapshots should always use ALUA, to prevent

a substantial performance hit.

Working with Snapshot Management The Snapshot Management page allows you to perform a variety of related tasks, using the following views:

•

Schedules View: Displays a list of schedules for LUNs and LUN groups, along with the type of schedule, and the number of snapshots created to date. You can select a LUN or LUN group from the list to view a list of snapshots in a table, or a calendar depicting the snapshot schedule.

•

LUN View: Displays a list of LUNs that have snapshots, along with the size of the LUN and the number of snapshots to date for each LUN. You can select a LUN and view a list of its snapshots, or a calendar depicting the snapshot schedule.

•

LUN Group View: Displays a list of LUN groups, along with the size of the LUN group and current number of snapshots. You can select a LUN group and view a list of its snapshots, or a calendar depicting the snapshot schedule. Note: A snap group is an object that represents a set of snapshots taken across a collection of LUNs that belong to a LUN group. Snapshots of each LUN in a LUN group are taken, then grouped as a snap group.

•

Table View: Displays a list of LUNs or LUN groups with snap shots or snap groups. All LUNs on the container are not shown.

•

Calendar View: Displays a calendar with showing scheduled snapshots by the month, week, day, or in a list.

Note: The LUN View panel (Manage > Manage Snapshots) only shows the LUNs and LUN groups that have snapshots and snap groups. It does not show all LUNs. To view all available LUNs, select Manage > Manage LUNs. To view all available LUN groups, select Manage > Manage LUN Groups.

84


535-0008-00 Rev 10

Manage

For more information on snapshots and how to create and manage them using the CLI, see CHAPTER 8, “Data Management” on page 253.

Creating Schedules This section demonstrates how to create immediate and scheduled snapshots of a LUN or LUN group. To create a one-time immediate snapshot, do the following: 1. Select 2. Click

Manage > Manage Snapshots.

Schedules and then click the Create icon.

The Create Snapshot/Schedule appears.

3. Select

Snapshot NOW, and enter a unique Snapshot Name in the text field followed by a description in the Description text field.

4. Select one of the following:

— LUN: Creates a snapshot of a selected LUN. — LUN Group: Creates a snapshot of a selected LUN group. The resulting snapshot is known as a snap group. 5. Select the LUN or LUN Group name from the drop-down list. The list shows all the available

LUNs and LUN groups in the container.

535-0008-00 Rev 10


85


6. Select the following check boxes, as necessary:

— Protected: When the number of snapshots exceeds a specified threshold, reclamation begins with the oldest unprotected snapshot. When this option is checked, a snapshot or snap group is protected from automatic free space reclamation. — Writable: By default, snapshots and snap groups are read-only. When is option is checked, the snapshot or snap group is writable. 7. Click

OK to create the snapshot, or Cancel to leave the operation without creating a snapshot.

To create a one-time scheduled snapshot, do the following: 1. Select 2. Click



3. Select

Scheduled. The dialog expands to show scheduling options.

4. Enter a unique

Snapshot Name in the text field, followed by a Description.

Note: A schedule is enabled is by default. If you disable a schedule, you must manually enable it again before it will begin taking automatic snapshots. Click the check box to disable the schedule; the check mark disappears. 5. Select one of the following:

— LUN: Creates snapshots of a selected LUN. — LUN Group: Creates snapshots of a selected LUN group. The resulting snapshot is known as a snap group. 6. Select

One-time snap. A check mark appears, designating that one snapshot will be taken at

the scheduled time. 7. Specify the day and time the snapshot will be taken, by doing the following:

— Select a Start Date from the pop-up calendar — Specify a Start Time in hours, minutes, AM or PM. 8. Select the following options, as necessary:

— Protected: When the number of snapshots exceeds a specified threshold, reclamation begins with the oldest unprotected snapshot. When this option is checked, a snapshot or snap group is protected from automatic free space reclamation. — Writable: By default, snapshots and snap groups are read-only. When this option is checked, the snapshot or snap group is writable.

86


535-0008-00 Rev 10

Manage

9. Click

OK to create the snapshot, or Cancel to leave the operation without creating a snapshot.

To create a scheduled for multiple snapshots, do the following: 1. Select 2. Click



The Create Snapshot/Schedule appears. 3. Select

Scheduled. The dialog expands to show scheduling options.

4. Enter a unique

Snapshot Name in the text field, followed by an optional Description.

Note: A schedule is enabled is by default. If you disable a schedule, you must manually enable it again before it will begin taking automatic snapshots. Click the check box to disable the schedule; the check mark disappears. 5. Select one of the following:

— LUN: Creates snapshots of a selected LUN. — LUN Group: Creates snapshots of a selected LUN group. The resulting snapshot is known as a snap group.

535-0008-00 Rev 10


87


6. Select the LUN or LUN Group name from the drop-down list, leave the

One-time snap check

box unchecked, and deselect the No End Date option. The dialog box expands. 7. Specify the following:

— Enter a Start Date in the text field, or select the date from the pop-up calendar. Then, set the Start Time using the drop-down menus. — Select the End Date (calendar) and End Time. 8. Specify the snapshot interval by entering a number in the Repeat every text box, and select a

frequency interval from the drop-down menu: Minute(s), Hour(s), Day(s) (default), Week(s), Month(s). 9. Specify the maximum number of snapshots to keep by entering a number in the Max. number

to keep text box. Then select the following check boxes, as necessary: — Protected: When the number of snapshots exceeds a specified threshold, reclamation begins with the oldest unprotected snapshot. When this option is checked, a snapshot or snap group is protected from automatic free space reclamation. — Writable: By default, snapshots and snap groups are read-only. When this option is checked, the snapshot or snap group is writable. 10.Click

OK to create the snapshot, or Cancel to leave the operation without creating a

snapshot.

Modifying and Deleting Schedules This section demonstrates how to modify existing schedules, and delete outdated schedules. To modify an existing schedule, do the following: 1. Select 2. Click


Schedules, then in the Schedules View select a schedule in the list. It highlights,

changing color. 3. Click

Edit.

The Edit Schedule dialog appears, showing information for the selected schedule.

88


535-0008-00 Rev 10

Manage

4. Make changes to the dialog option fields, as necessary.

5. Click

OK to apply the changes, or Cancel to leave the task without changing to the existing

schedule. To delete a schedule, do the following: 1. Select 2. Click


Schedules, then in the Schedules View select a schedule in the list. It highlights,

changing color. 3. Click

Delete.

The Delete Schedule dialog appears. 4. Select the check box to delete all snapshots created by the schedule, or accept the unchecked

default to delete only the schedule. Then, click OK to make the deletion or Cancel to abort the task.

535-0008-00 Rev 10


89


Viewing Schedules This section demonstrates how to view a list of schedules, associated snapshots, and exported snapshots. To view a list of schedules and associated snapshots, do the following: 1. Select 2. Click


Schedules.

The Schedule View panel on the left displays a list of schedules. The One-Time-Grouped schedule is selected by default, and is a collection of one-time scheduled snapshots. The schedule is highlighted, and a list of snapshots and snap groups is shown on the right. Note: The One-Time-Grouped schedule can contain snapshots and snap groups. All other schedules contain either snapshots or snap groups, but not both. 3. To view the snapshots and snapgroups for another schedule, select the schedule in the

Schedule View. The schedule changes color (highlights). The associated snapshots and snapgroups appear in the panel on the right.

To view a list of exported snapshots, do the following: 1. Select 2. Click


Schedules, then select a schedule from the list.

The schedule changes color (highlights). The associated snapshots and snap groups appear in the panel on the right. 3. Select a snapshot from the list on the right. The selected snapshot changes color.

90


535-0008-00 Rev 10

Manage

4. Click the number in the Export column on the far right. A pop-up dialog appears, showing a list

of exported snapshots.

5. To search for a target or initiator, enter a text string in the 6. Click

Search field and press Enter.

Close to return to the Snapshot Management page.

Creating LUN Snapshots This section demonstrates how to create an immediate snapshot of a selected LUN.

Note: The only LUNs listed in the LUN View panel (Manage

> Manage Snapshots) are LUNs that have snapshots. It does not show all available LUNs in the container. To view all available LUNs, select Manage > Manage LUNs. For information on how to set up a snapshot schedule, see Creating Schedules on page 85. To create an immediate snapshot of a LUN, do the following: 1. Select


2. Click

LUNs to display the LUN View panel, and select a LUN from the list. The selected LUN highlights, changing color. The snapshots associated with the LUN appear in the panel to the right.

3. Click

535-0008-00 Rev 10

+ to create a new snapshot.


91


The Create Snapshot/Schedule dialog appears.

4. Leave

Snapshot NOW selected, and enter a Snapshot Name. You can also enter an optional Description.

5. Leave

LUN selected, and select the LUN Name from the drop-down list, if not already

selected. 6. Select the following check boxes, as necessary:

— Protected: When the number of snapshots exceeds a specified threshold, reclamation begins with the oldest unprotected snapshot. When this option is checked, a snapshot or snap group is protected from automatic free space reclamation. — Writable: By default, snapshots and snap groups are read-only. When this option is checked, the snapshot or snap group is writable. 7. Click

OK to create the snapshot, or Cancel to leave the task without creating a snapshot.

Creating LUN Group Snapshots A snapshot of a LUN group consists of snapshots of each LUN in the LUN group, combined into a group called a snap group.

92


535-0008-00 Rev 10

Manage

Note: The only LUN Groups listed in the LUN Group View panel (Manage > Manage Snapshots) are those that have snapshots, not all the available LUN Groups in the container. To view all LUN groups in the container, select Manage > Manage LUN Groups. For information on

how to set up a snapshot schedule, see Creating Schedules on page 85.

To create a snap group, do the following: 1. Select 2. Click


LUN Groups, and then click +.

The create Snapshot/Schedule dialog appears. 3. Leave Snapshot NOW selected and enter a Snapshot Name. You can also enter an optional

Description. 4. Leave

LUN Group selected and choose a name from the LUN Groups Names drop-down

list. 5. Select the following check boxes, as necessary:

— Protected: When the number of snapshots exceeds a specified threshold, reclamation begins with the oldest unprotected snapshot. When this option is checked, a snapshot or snap group is protected from automatic free space reclamation. — Writable: By default, snapshots and snap groups are read-only. When this option is checked, the snapshot or snap group is writable.

6. Click

535-0008-00 Rev 10

OK to create the snapshot, or Cancel to leave the task without creating a snapshot.


93


Viewing Snapshots in a Snap Group To view a list of snapshots in a snap group, do the following: 1. Select 2. Click


Schedules, then select a a LUN group schedule from the list on the left.

The LUN group schedule changes color (highlights) and the snap groups appear in the panel on the right. 3. Select a snap group from the list on the right. The selected snapshot changes color.

4. In the # column on the far right, click the number link. The snapshots belonging to the snap

group appear in the panel below.

5. Click

Hide in the upper right corner of the snap group snapshots panel to hide the list of snapshots from view.

94


535-0008-00 Rev 10

Manage

Editing LUN and LUN Group Snapshots This section demonstrates how to edit LUN and LUN group snapshots from the Snapshot Management page.

Note: The only LUNs or LUN Groups listed in the LUN View or LUN Group View panels (Manage > Manage Snapshots) are those that have snapshots, not all the available LUN Groups in the container. To view all available LUNs and LUN groups in the container, select Manage > Manage LUNs or Manage > Manage LUN Groups. For information on editing snapshot schedules, see

Modifying and Deleting Schedules on page 88.

To edit a snapshot of a LUN or LUN group, do the following: 1. Select


2. Click

LUN or LUN Groups, and then select the LUN or LUN Group name from the list on the left so that it highlights, changing color. A list of snapshots associated with the selected LUN or LUN Group appears in the panel on the right.

3. In the right panel, select the snapshot you want to edit so that it highlights, changing color.

Then click the Edit icon.

The edit dialog appears.

535-0008-00 Rev 10


95


4. Make the necessary changes to the following options:

— Snapshot or Snap group Name Note: A snapshot or snap group can only be renamed if they are not exported.

— Description — Protected — Writable

5. Click

OK to apply the changes, or Cancel to leave the operation without making any changes.

Exporting LUN and LUN Group Snapshots This section demonstrates how to export a LUN snapshot or snap group to selected initiators, initiator groups, and ports.

Note: The only LUNs or LUN Groups listed in the LUN View or LUN Group View panels (Manage > Manage Snapshots) are those that have snapshots, not all the available LUN Groups in the container. To view all available LUNs and LUN groups in the container, select Manage > Manage LUNs or Manage > Manage LUN Groups.

To export a snapshot of a LUN or LUN group, do the following: 1. Select 2. Click


LUNs or LUN Groups.

3. Select a LUN or LUN Group name in the list on the left so it highlights, changing.

A list of snapshots associated with the selected LUN or LUN Group appears in the panel on the right. 4. In the right panel, select the check box for the snapshot you want to export. A check mark

appears.

96


535-0008-00 Rev 10

Manage

5. Click the

Export icon.

The Add Export dialog appears. All Initiators and All Ports are the defaults.

6. Do the following, as necessary:

— To export the snapshot or snap group to All Initiators and All Ports, go to step 5. — To export to selected initiators and initiator groups, click Specific Initiator Groups and/ or Initiators, and select the Initiator Groups and/or Individual Initiators from the lists. Press Ctrl to select multiple initiators or initiator groups. — To export to selected ports, click Specific Ports, and select the ports from the list. Press Ctrl to select ports. — LUN ID: Auto (automatic assignment of a LUN ID) is selected by default. To specify a unique LUN ID, click Value and enter a number in the text box. Note: LUN IDs must be unique within the container.

7. Click

OK to export the snapshot to the selected initiators and ports, Or click Cancel to exit the operation without facilitating an export.

535-0008-00 Rev 10


97


Removing Exports from LUN and LUN Group Snapshots This section demonstrates how to remove all exports from a LUN snapshot or LUN group snap group.


To remove all exports from a LUN or LUN group snapshot, do the following: 1. Select 2. Click


LUNs or LUN Groups.

3. Click the check box to the left of a LUN or LUN Group snapshot (that you want to remove the

exports from) in the list on the right, so a check mark appears. Note: Only checked snapshots are affected by this operation. 4. Click

Remove Exports. A confirmation dialog appears.

5. Click

OK to remove all exports, or Cancel to leave the operation without removing the exports.

Rolling Back a LUN or LUN Group from a Snapshot This section demonstrates how to roll back a LUN or LUN group from a snapshot. For more information on the roll back process, see Rolling Back the Contents of a Snapshot on page 267. The only LUNs or LUN Groups listed in the LUN View or LUN Group View panels (Manage > Manage Snapshots) are those that have snapshots, not all the available LUN Groups in the container. To view all available LUNs and LUN groups in the container, select Manage > Manage LUNs or Manage > Manage LUN Groups.

Note: A LUN or LUN group has to be unexported prior to performing a rollback. For more

information, see Exporting and Unexporting LUNs in a LUN Group on page 79.

To roll back a LUN or LUN group from a snapshot, do the following: 1. Select 2. Click


LUN or LUN Groups.

3. Select the LUN or LUN group (you want rolled back) from the list of snapshots on the left.

98


535-0008-00 Rev 10

Manage

4. Select the LUN or LUN Group snapshot (whose contents you want to roll back to the LUN or

LUN group) from the list on the right, so that it changes color (highlights). 5. Click

Rollback. A confirmation dialog appears.

6. Click

OK to remove all exports, or Cancel to leave the operation without rolling back the LUN

or LUN group.

Deleting LUN and LUN Group Snapshots This section demonstrates how to delete LUN and LUN Group snapshots. Review the following guidelines before you begin deleting snapshots:

•

Snapshots that belong to a snap group cannot be deleted individually. However, you can delete the entire group of snapshots.

•

Snapshots that have been exported cannot be deleted. You must unexport the snapshot first, then it can be deleted.

•

Protected snapshots cannot be deleted, unless you have Admin user privileges. However, it is recommended that you unprotect the snapshots first, then delete them.


To delete a LUN or LUN group snapshot, do the following: 1. Select 2. Click


LUN or LUN Groups.

3. Select a LUN or LUN Group from the list on the left. A list of associated snapshots appears in

the panel on the right. 4. Click the check box to the left of a snapshot (you want to delete) in the list on the right, so that

a check mark appears. Note: Only checked snapshots are affected by this operation.

535-0008-00 Rev 10


99


5. Click the

Delete icon.

A confirmation dialog appears.

6. Click OK to delete the selected snapshot, or Cancel to leave the operation without deleting the

snapshot.

Viewing Schedule Calendars Calendar View provides an at-a-glance map of current schedules, with each schedule appearing as a different color across the following formats:

•

month—shows a current month planner, with schedules appearing on days snapshots are taken

• • •

week—shows a current week planner, with schedules appearing on days snapshots are taken day—shows a current day planner, with schedules appearing at the time they are taken list—shows a list of scheduled snapshots for the current day

Note: Calendar View reflects current schedules only. When you change or delete a schedule, the

Calendar View changes accordingly.

To view a calendar display of schedules, do the following: 1. Select


2. At the far right of the Snapshot Management page, click

Calendar. The Calendar View appears, showing the default month view. The current day is highlighted.

100


535-0008-00 Rev 10

Manage

3. To display a week-planner schedule view, click week. Schedules appear in the time slot of the

day or days on which the snapshots are taken. The current day is highlighted.

4. To display a day-planner schedule view, click

day. Schedules appear in the time slot in which

the snapshots are taken.

535-0008-00 Rev 10


101


5. To display a list of schedules for the current day, click

6. To return to the Snapshot Management page, click

list.

Table View.

Space Management You can monitor and manage the available space on containers for regular (thick) LUNs, thin LUNs, LUN groups, snapshots, and snap groups. This section covers the following topics:

• • •

Managing Array Space Usage on page 102 Managing Container Space Usage on page 103 Managing LUN Space Usage on page 104

Managing Array Space Usage This section demonstrates how to retrieve an at-a-glance view of space usage for an array. For information on how to reclaim free space, see Rules for Reclaiming Free Space on page 283 and Setting Usage Thresholds on page 283. To view array space usage, do the following: 1. Select

Manage > Space Management.The Space Management page appears.

2. Review the left-most panel for the following specific container information:

— Container Name — Number of LUNs (in the container) — HA (High Availability) Status

3. In the center panel, review container space usage in the following areas:

— — — — — — —

102

At-a-glace bar graph showing space usage Thick LUNs, space used (actual Gigabytes) and percentage of container space used Thin LUNs, space used (Gigabytes) and percentage of container space used Snapshots, space used (Gigabytes) and percentage of container space used Reserved, space (Gigabytes) reserved for internal metadata management Free, available free space (Terabytes) and percentage of container space Total space in the container


535-0008-00 Rev 10

Manage

4. Review the right-most panel for the following threshold information:

— Provision Thresholds: Alert at: A warning occurs when the provisioned space reaches or goes above this percentage Limit: Prevents any space provisioning above this percentage — Used Space Thresholds Alert at: A warning occurs when space usage reaches or goes above this level Reclaim: Space reclamation starts when space usage goes above this level

5. To change a threshold, click the

Edit icon, enter a new Threshold Percentage value in the

text field, and click OK.

Managing Container Space Usage This section demonstrates how to review space usage for a container. For information on how to reclaim free space, see Rules for Reclaiming Free Space on page 283 and Setting Usage Thresholds on page 283. To view space usage for a container, do the following: 1. Select 2. Click

Manage > Space Management.

Container View.

The Container View panel appears.

535-0008-00 Rev 10


103


3. Review the top-most pane for information on maximum usable capacity, allocated size,

maximum provisioned capacity, and current provisioned size. 4. Review the left Allocated pane for a breakdown of the physical space (Terabytes and

Gigabytes) used for LUNs, snapshots, and snap groups, as well as the amount available free space. The graph in the bottom half of the pane illustrates the data. 5. Review the right Provisioned pane for a breakdown of the logical space (Terabytes and

Gigabytes) provisioned for LUNs and snapshots, as well as the remaining available free space. The graph in the bottom half of the pane illustrates the data.

Managing LUN Space Usage This section demonstrates how to review space usage for a selected LUN. For information on how to reclaim free space, see Rules for Reclaiming Free Space on page 283 and Setting Usage Thresholds on page 283.

Note: The LUN View panel only shows LUNs with snapshots, not all LUNs in the container.

To view space usage for a LUN, do the following: 1. Select 2. Click

Manage > Space Management.

LUN View, then select a LUN name from the list on the left so that it highlights.

The details of the selected LUN appear in the panel on the right, as shown in the following example.

104


535-0008-00 Rev 10

Monitor

3. View a list of LUN names, the Type, Provisioned Size (Gigabytes), and Size Used (Gigabytes)

in the left pane. 4. Enter a text string in the

Search field to search for a LUN.

5. Sort the contents of a column by clicking the up arrow (ascending) or down arrow

(descending). 6. To view details of a specific LUN, select the LUN in the list (in the left pane) so it highlights,

changing color. The details for the LUN appear in the LUN Details pane (on the right). 7. Enter a text string in the Search field to search for a particular snapshot of the selected LUN.

Monitor The Monitor function provides an at-a-glance view of the health and performance of cluster nodes, and the ability to analyze historical data and real time trends. LUN monitoring provides container level statistics both on a per-node basis, as well as aggregated. Target monitoring provides port level statistics. The following table provide a list of tasks and their associated functions. Click a hyperlink to jump directly to the task. Function

Monitor Overview Monitor LUNs Monitor LUN Groups Monitor Targets

535-0008-00 Rev 10

Tasks

• •

Customizing the Output Display on page 106

• •

Searching for a LUN on page 110

• •

Searching for a LUN Group on page 111

• •

Selecting a Target Port on page 112

Monitoring Container, LUN, and Port Data on page 108 Analyzing Data for a LUN on page 110 Analyzing LUN Group Data on page 111 Analyzing Data for a Target on page 112


105


Monitor Overview The Overview function provides a high-level view of real-time or historical data. You can customize how you view the data, choosing between line and area graphs, time trends, as well as the type of metrics used to formulate the data. You also have the option to view data for a container, LUN, or port.

•

Real Time Performance Trend—Shows the last five minutes of performance trends for the selected metric. New real time values are appended into the graph every ten seconds.

•

Current Value—Shows real time data at this time. The graph time is listed at the bottom of the panel.

Customizing the Output Display The following task demonstrates how to customize the data display to suite your needs. You can specify the visual graph you prefer, the time trend, and the applied metrics. You can also choose to view the aggregate data (A+B) from both Memory Gateway nodes, or view data for the nodes separately (A, B). If you select the A only or B only option, only the data for that particular node is shown. To customize the data output display, do the following: 1. Select 2. Click

Monitor > Overview. The Monitoring page appears.

Settings & Download. The panel expands to show the following options.

3. Select the

Type of information to display from the drop-down list.

•

General—Monitors data for all ports.

•

LUN—Monitors data for a specific LUN.

•

Fibre Channel—Monitors data for Fibre Channel ports.

•

InfiniBand—Monitors data for InfiniBand ports.

For more information, see Monitoring Container, LUN, and Port Data on page 108. 4. Select a Metric display type from the drop-down list, as shown in the following example.

•

A+B shows aggregated data for both Gateway nodes.

•

A, B shows data separately for both Gateway nodes.

•

A only shows data for the Gateway master node. B only shows data from the standby node.

The results appear in the Current Value panel on the right. 106


535-0008-00 Rev 10

Monitor

5. Choose an option from the drop-down list to customize the graph display. The Real-Time

Performance Trend display changes accordingly.

6. To specify the real-time performance trend, at the top of the left-most panel choose one of the

following options:

•

Last 1 H: Shows data averaged every one minute.

•

Last 24 H: Shows data averaged over the last 24 hours.

7. To view detailed information for a point in time, move the cursor over a point on the graph. A

pop-up dialog appears, as shown in the following example.

8. Click

535-0008-00 Rev 10

Settings & Download again to hide the panel.


107


Monitoring Container, LUN, and Port Data You can analyze data in different containers, LUNs, and across different ports. This section demonstrates how to perform each of these tasks. To monitor a specific container, do the following: 1. Select 2. Click


Settings & Download to expand the panel.

3. From the Type drop-down list, choose

General (default setting).

4. (Optional) To save a snapshot of performance data, do the following: a. Click the

Download check box. b. Choose a Time Interval from the drop-down list. c. Click Download as CSV. Data for the specified time interval is saved as a CSV file.

5. (Optional) View detailed information on the container by clicking the blue information (i) icon in

the Current Value panel or next to the Container drop-down list.

6. (Optional) Modify the output display, as described in Customizing the Output Display on page

106. To monitor a specific LUN, do the following: 1. Select 2. Click



3. From the Type drop-down list, choose

LUN, then select a name from the LUN drop-down list

on the right.

108


535-0008-00 Rev 10

Monitor

4. (Optional) View detailed information on the LUN by clicking the blue information (i) icon. 5. Select a Metric from the drop-down list. Click the blue information (i) icon for a description of

each option. The Current Value panel changes to display the data for the selected metric. 6. (Optional) Modify the output data display, as described in Customizing the Output Display on

page 106. To monitor a port, do the following: 1. Select 2. Click



3. Choose a port from the

Type drop-down list, then choose a port name from the drop-down list

that appears to the right. Note: The General option monitors data for all ports. A port drop-down list does not appear when this option is selected. 4. (Optional) To view detailed information about the port, click the information (i) icon next to the

port name.

5. Choose a Metric from the drop-down list, as available. Some port selections only provide one

option. 6. (Optional) Modify the output data display, as described in Customizing the Output Display on

page 106.

535-0008-00 Rev 10


109


Monitor LUNs You can monitor the performance of selected LUNs in a container. The LUN Monitoring page allows you to search for and view real-time statistics or historical trends for selected LUNs.

Searching for a LUN This section demonstrates how to search for a specific LUN or set of LUNs. To search for a specific LUN, do the following: 1. Select

Monitor > Monitor LUNs. The LUN Monitoring page appears.

2. Enter a string in the search field and press

Enter. Results matching the search string appear

in the panel on the left.

3. Select the LUN so that it highlights. Real-time data for the LUN appears in the panels on the

right.

Analyzing Data for a LUN You can view and analyze the following performance data types for a selected LUN:

• • •

Read/Write Bandwidth Read/Write IOPs Read/Write Latency

To analyze the performance for a specific LUN, do the following: 1. Select

Monitor > Monitor LUNs.

2. To find a specific LUN, follow the instructions described in Searching for a LUN on page 110,

then select the LUN so that it highlights. 3. Select one of the following time intervals:

— Real Time Displays data as it happens, in real time — 1 H Displays aggregated data for the last hour — 24 H Displays aggregated data for the last 24 hours 4. Choose a graph type from the drop-down list: Line, Area, Stacked Line, or Stacked Area. 5. Rest the mouse cursor over a point on a graph to view detailed information for that specific

point in time.

110


535-0008-00 Rev 10

Monitor

Monitor LUN Groups You can monitor the performance of selected LUN groups in a container. The LUN Group Monitoring page allows you search for LUN groups and view real-time statistics or historical trends for selected LUN groups.

Searching for a LUN Group This section demonstrates how to search for a specific LUN or set of LUNs. To search for a specific LUN, do the following: 1. Select

Monitor > Monitor LUN Groups. The LUN Monitoring page appears.

2. Enter a string in the search field and press

Enter. Results matching the search string appear

in the panel on the left. 3. Select the LUN group so that it highlights. Real-time data for the LUN group appears in the

panels on the right.

Analyzing LUN Group Data You can view and analyze the following performance data types for a selected LUN group:

• • •

Read/Write Bandwidth Read/Write IOPs Read/Write Latency

To analyze the performance for a specific LUN, do the following: 1. Select

Monitor > Monitor LUN Groups.

2. To find a specific LUN, follow the instructions described in Searching for a LUN Group on page

111, then select the LUN Group so that it highlights. 3. Select one of the following time intervals:

— Real Time Displays data as it happens, in real time — 1 H Displays aggregated data for the last hour — 24 H Displays aggregated data for the last 24 hours 4. Choose a graph type from the drop-down list: Line, Area, Stacked Line, or Stacked Area. 5. Rest the mouse cursor over a point on a graph to view detailed information for that specific

point in time.

Monitor Targets You can analyze performance data for target ports in time statistics, as well as using historical trends.

Note: Fibre Channel data is shown in frames per second. InfiniBand data is shown in packets per

second.

535-0008-00 Rev 10


111


Selecting a Target Port The target ports configured for your system are easily accessible from the Monitor Targets page. To select a target port for monitoring, do the following: 1. Select

Monitor > Monitor Targets. The Target Port Monitoring page appears. Statistics for each target port are shown in the left panel, with real time data in a graph format on the right.

2. Select a port name from the list on the left so that it highlights. The graphs on the right side of

the window change to display data for the selected target.

Analyzing Data for a Target You can analyze target data in real-time or as historical trends, and choose from a variety of display formats. To analyze data for a selected target port, do the following: 1. Select

Monitor > Monitor Targets. The Target Port Monitoring page appears.

2. Choose a target port, as described in Selecting a Target Port on page 112. 3. Select one of the following time intervals for data analysis:

— Real Time Displays data as it happens, in real time — 1 H Displays aggregated data for the last hour — 24 H Displays aggregated data for the last 24 hours 4. Choose the graph type from the drop-down list on the right.

•

Line

•

Area

•

Stacked Line

•

Stacked Area

The following example shows port data using a Stacked Area graph.

112


535-0008-00 Rev 10

Administration

Administration The Administration page of the Violin Web Interface provides a single point of contact where you can perform a wide range of Memory Gateway administration tasks. The following table provide a list of tasks and their associated functions. Click a hyperlink to jump directly to the task. Function Page

Cluster Admin

Tasks

• • • • • • • • •

Viewing Software Version and Boot Images on page 114

• • • • • • •


• • • • •


Tools

•


Information

• • •


Network

Notification

535-0008-00 Rev 10

Rebooting or Shutting Down a Memory Gateway on page 114 Changing and Saving Configuration Files on page 115 Viewing Gateway Node Details on page 115 Viewing Alerts on page 116 Searching for and Viewing Logs on page 117 Generating and Downloading Combined Log Files on page 118 Managing User Accounts on page 119 Viewing Node Versions on page 122 Changing the Cluster Management Address on page 123 Modifying Global Default Gateway Settings on page 123 Viewing and Modifying DNS Settings on page 124 Viewing and Modifying NTP Settings on page 125 Viewing and Modifying Web Admin Settings on page 126 Viewing Gateway Node Details on page 115 Managing E-mail Settings on page 129 Managing E-mail Notifications on page 132 Managing SNMP Settings on page 134 Managing SNMP Traps on page 136

Accessing a Full PDF Version of this Guide on page 141 Viewing System, License, and Copyright Information on page 142


113


Cluster Admin Cluster administration includes configuration and management tasks for the nodes in a cluster. A cluster consists of the two Memory Gateway nodes, a master node and a standby. The Master Memory Gateway is the current master node in the cluster. The master node works together with a standby node to manage the availability of the cluster.

Viewing Software Version and Boot Images The Software section of the Cluster Admin page shows the current software version running on the Memory Gateways, as well as the boot images. Two boot images are installed for each Memory Gateway, each in a separate partition. To view the software version and choose a boot image, do the following: 1. Select

Administration > Cluster Admin > Reboot.

2. To view the version of software running on the Memory Gateway, look at the

Software

Version section of the Software panel. 3. To view the installed boot images, look at the

Installed Images section of the Software panel.

Rebooting or Shutting Down a Memory Gateway You can reboot or shut down a Memory Gateway through the Violin Web Interface.

•

Shutdown—Stops a Memory Gateway node or removes the node from the cluster for planned maintenance.

•

Reboot—Restarts a Memory Gateway node, clearing its data but retaining the existing configuration settings. In a cluster, the standby Memory Gateway takes over while the previously active master node reboots.

Caution: Rebooting a Memory Gateway node may cause a temporary decline in overall performance.

To reboot or shut down a Memory Gateway node, do the following: 1. Select


The Cluster Administration page displays information on the current software version, the installed software image, and configuration files. Two boot images are installed for each Memory Gateway in separate partitions. Note: Rebooting the system will take several minutes to complete. 2. Do one of the following:

114


535-0008-00 Rev 10

Administration

— To reboot the system, click Reboot. In a cluster configuration, the standby Memory Gateway takes over while the master reboots. — To shut down the system, click Shutdown. In a cluster configuration, the standby Memory Gateway takes over when the master is shut down. A pop-up dialog appears. 3. Click

OK to continue with the reboot process, or Cancel to quit without rebooting or shutting down the system.

WARNING! After the system is shut down, you will have to manually power it back on.

Changing and Saving Configuration Files You can save the current Memory Gateway configuration to a specified file, apply a previously saved configuration to the cluster (whether a local file or on a remote server), and delete old configuration files. The currently active configuration is marked as Active in the configuration table. The Status column indicates which configuration file is active. The active file is backed up as .bak. To save, apply, or delete a Memory Gateway configuration file, do the following: 1. Select


2. To apply a previous configuration to the Memory Gateway, select the check box next to the

Config Name and click Apply.

3. To save the current Memory Gateway configuration, enter a unique name in the

Save

configuration as field, and click Save. 4. To delete a configuration, select the check box next to the

Config Name and click Delete.

Viewing Gateway Node Details The Gateways page lists all the Memory Gateways (nodes) in the cluster, their status (online/ offline), their role (master node, standby node), and notification of unsaved changes To view Memory Gateway cluster node details, do the following: 1. Select

Administration > Cluster Admin > Gateways.

The Gateways page displays a list of active nodes.

535-0008-00 Rev 10


115


2. To change the number of nodes displayed per page, select a number from the

Display drop-

down list and click Apply. The range of display alerts appears between the Prev and Next hyperlinks. 3. To view detailed information on a particular node, click the Gateway node link in the list.

The View Gateway page shows the internal and external IP addresses, the node’s role and status, uptime, software version, and the name of its installed image. The page also shows the amount of DRAM installed on the node.

4. Click

Show Gateway List to return to the list of Gateway nodes.

Viewing Alerts The Alerts page shows a listing of current and past system events that triggered an alert. Select the types of alerts to display (informational, warning, or error) and click the APPLY button. To view cluster alerts, do the following: 1. Select

Administration > Cluster Admin > Alerts.

2. To change the types of alerts that appear in the list, deselect or select the check box next to

the Show icon: Info, Warning, Error. A check mark indicates that the alert type is selected to appear in the list. 3. Click the

Next to page forward through the list and Prev to go back.

4. To change the number of alerts displayed on a page, choose a number from the Display drop-

down list and then click Apply.

116


535-0008-00 Rev 10

Administration

Searching for and Viewing Logs The Web Interface allows you to easily view a listing of logged events. You can choose between viewing current and historical logs in paginated format. Log files are rotated once a day.

Note: You can only view logs for the current master gateway using the Violin Web Interface.

To view logs, do the following: 1. Select

Administration > Cluster Admin > Logs.

2. Select a

Log File from the drop-down list, and click Apply. The current log is the default.

3. To navigate to the desired log page, do any of the following:

— Click Next (>) to page forward through the log files and Prev (>) to jump to the last page, and the double lessthan arrows () last available page in the

log and scroll to the bottom. Note: The Log File field displays the top file in the drop-down list (“Current log”). It does not

display the name of the log file that you are viewing, which appears directly below the Log File field.

535-0008-00 Rev 10


117


Generating and Downloading Combined Log Files You can generate a user-selected set of log files from Memory Gateways and Memory Arrays, then download the file onto a local machine. To generate and download a combined log file, do the following: 1. Select

Administration > Cluster Admin > Logs.

2. In the Download the Combined Log Files panel, choose one of the following options from the

Generate the Combined Log File drop-down list: — All (Gateways + Arrays) generates a combined log file for all the Memory Gateways and Memory Arrays. — All Gateways generates a combined log file for all the Memory Gateways. — All Arrays generates a combined log file for all the Memory Arrays. — mg-a generates a combined log file for the mg-a Memory Gateway. — mg-b generates a combined log file for the mg-b Memory Gateway. — acm-a generates a combined log file for the acm-a Memory Array. — acm-b generates a combined log file for the acm-b Memory Array. 3. Click

Start.

The generation generally takes a few minutes to complete. The Generation Status line apprises you of the progress. When the combined log file is generated, the Download button becomes active.

4. Click

Download to save the newly generated combined log file to the local machine.

When active, the Download button has a mouse-over ToolTip that shows the log file name, which includes a time stamp showing when the file was generated. The combined log files are contained in a single .tar file.

118


535-0008-00 Rev 10

Administration

Managing User Accounts User Management displays the name, role, email address, and password status of each user account for the cluster. This section demonstrates how to perform the following tasks:

• • • •

Deleting a User Account on page 121 Changing an Account Password on page 120 Editing a User Account on page 120 Deleting a User Account on page 121

Understanding User Roles and Privileges

A user’s assigned role determines the level of privileges they have in accessing and administering the cluster. Some of parameters are only visible or configurable by users with certain roles (privileges). The available roles and their privileges are as follows: User Role

Privileges

admin

A user who can perform all administration and configuration tasks.

monitor

A user who can only access pages that show status information for monitoring purposes.

Table 2.12 User Roles and Privileges Adding a User and Specifying a Role

An Admin user can add new users and define their account privileges. To add a new user account and define the access privileges, do the following: 1. Select

Administration > Cluster Admin > Users.

2. At the bottom of the page, click Add

New User. The Add New User dialog appears.

3. Specify the following account information:

• •

User Name: Enter a user name for login credentials. Role: Select a role from the drop-down list, admin or monitor. For more information, see Understanding User Roles and Privileges on page 119.

4. Click

OK to add the new account, Cancel to abort the action, or Reset to clear the fields and start over.

535-0008-00 Rev 10


119


Changing an Account Password

An Admin user can set or reset the passwords and define access privileges for other user accounts. To add a new user account and define the access privileges, do the following: 1. Select


2. Next to the user name, click

Change Password. The Change Password dialog appears.

3. Specify the following information:

• • •

User Name: Enter the user name for the login credential. Password: Enter the new password. Confirm: Re-enter the new password.

4. Click

OK to accept the change, Cancel to abort the action, or Reset to clear the fields and start over.

Editing a User Account

An Admin user can change an existing user account, modifying personal and contact information. To edit an existing user account, do the following: 1. Select 2. Click


Edit for the desired user. Edit User dialog appears.

3. Change the following account information, as necessary:

• • • •

120

Role: Select a role from the drop-down list, admin or monitor. For more information, see Understanding User Roles and Privileges on page 119. First Name: Enter the user’s first name.

Last Name: Enter the user’s last name. Email Address: Enter the user’s email address.


535-0008-00 Rev 10

Administration

4. Click

OK to accept the changes, Cancel to abort the action, or Reset to clear the fields and start over. If you accept the changes, another dialog appears.

5. Click

OK in the confirmation dialog to save the changes. The results appear in the User Management panel.

Deleting a User Account

An Admin user can delete the accounts of other users. To delete a user account, do the following: 1. Select 2. Click


Delete next to the account name you want to remove. A confirmation dialog appears

.

3. Click

Yes to confirm you choice, or No to cancel the action. If you click Yes, the user account is deleted from the list.

535-0008-00 Rev 10


121


Viewing Node Versions The Versions page shows the software version running on each Memory Gateway and identifies the current boot image. It also shows how long each Memory Gateway has been running (Uptime) and identifies the boot Memory Gateway and next Memory Gateway to be booted (Boot/Next). To view version details for a Memory Gateway node, do the following: 1. Select Administration > Cluster Admin > Versions. The Versions page appears with a list of

active cluster nodes and the following information for each: — — — — —

Gateway ID: Node name Uptime: How long each node has been running Version: Software version running on the node Boot/Next: The boot Memory Gateway and next Memory Gateway to be booted Violin Cluster ID String: Current boot image

2. View more details on a node, click the Gateway ID link on the left. The View Gateway page

appears, as described in Viewing Gateway Node Details on page 115.

Network The Network Web Interface pages provide network configuration information for the cluster, and allows you to manage the cluster address and global default gateway. The tasks you can perform on these pages are intended for network administrators, and require Admin user access privileges. The cluster management address (also known as the virtual IP address, or VIP) is the IP address for the Memory Gateway master.

Viewing Current Network Settings The Network Settings page allows you to view and modify network settings from one central location. To view current cluster network settings, do the following: 1. Select

Administration > Cluster Admin > Network > Network Settings. The Network

Settings page appears. 2. Verify the

Cluster Management Address and Network Mask Length.

The IP address for cluster management is set as part of the initial out-of-the box configuration. You can change the cluster management IP address and network mask length in this section, but it is not recommended. Consult with your network architect before changing these values.

122


535-0008-00 Rev 10

Administration

3. Review the

Bonding Mode setting.

This refers to Cluster Interface Bonding. For more information on bonded interfaces, see Network Bond Commands on page 315. 4. Review the

Routing settings.

This is the address for the default global gateway. You can enable and disable the gateway and set a new address for the default gateway. 5. Review the cluster

Static and Dynamic Routes.

This is the routing information for the cluster, including destination, mask, gateway, whether the node is active or static, and the interface.

Changing the Cluster Management Address You can change the cluster management address (also known at the virtual IP address or VIP) from the Network Settings page. In most cases, it is recommended that you change the cluster management address through a serial console using the CLI. However, there are few situations where changing the cluster management address through the Web Interface is advised, one being in preparation for a physical move of the devices.

Note: Consult with your network administrator before changing the cluster management address.

If you change this address through the Web Interface, you lose connectivity with the cluster.

To change the cluster management address, do the following: 1. Select


Settings page appears. 2. In the Cluster Management Address panel, enter an 3. Enter the 4. Click

IP Address in the text field.

Network Mask Length in the text field.

Apply.

Modifying Global Default Gateway Settings The Network Settings page allow you to specify the Global Default Gateway and then enable or disable its use. To modify the global default gateway, do the following: 1. Select


Settings page appears. 2. To disable the current global default gateway, click

Disable. To re-enable the default gateway,

click Enable. 3. To change the global default gateway, enter the new default gateway address in the text field

and click Set Default Gateway.

535-0008-00 Rev 10


123


Viewing and Modifying DNS Settings You can view domain name server (DNS) configurations, add DNS servers, as well as add and remove domain names from the DNS Settings page.

Note: If one or more IP addresses for DNS servers were set up as part of the initial out-of-the-

box setup, they are listed on this page.

To view configured DNS settings, do the following: 1. Select

Administration > Network > DNS Settings.

2. Go to the DNS Servers panel at the top of the window to view information on the DNS servers

that were configured as part of the initial out-of-the-box setup.

To add and modify name servers, doe the following: 1. Select


2. To modify a configured name server, go to the Add or Modify Name Servers panel and change

the Primary DNS IP address displayed in the text box. 3. To add a second name server, enter an IP address in the

Secondary DNS IP address text

box. 4. To add a third name server, enter an IP address in the

5. Click

Tertiary DNS IP address text box.

Apply.

To add and remove domain names, do the following: 1. Select


2. To remove an existing domain name, go to the

Domain Names panel and select the domain

name from the list. The selected name highlights.

124


535-0008-00 Rev 10

Administration

3. Click

Remove Selected.

4. To add a new domain name, enter the fully qualified domain name in the text field in the Add

New Domain Name panel, then click Add Domain Name.

Viewing and Modifying NTP Settings The Network Time Protocol (NTP) is a networking protocol for synchronizing the clocks of computer systems over data networks. You can enable or disable NTP synchronization and add or remove an NTP server through the NTP Settings page. When you initially go to the NTP Settings page, information for the NTP server that was configured in the out-of-box setup is shown. To enable or disable NTP time synchronization, do the following: 1. Select

Administration > Network > NTP Settings.

2. To enable NTP, click the

Enable NTP Time Synchronization check box. A check mark appears to show it is enabled. Then click Apply.

3. To disable NTP time synchronization, click the check marked

Enable NTP Time

Synchronization check box. The check mark disappears to show that it is disabled. Then click Apply. To enable, disable, or remove an NTP server, do the following: 1. Select


2. To enable an NTP server, click the check box next the server name in the NTP Servers panel.

Then, click Enable Server.

535-0008-00 Rev 10


125


3. To disable an NTP server, click the check box next server name in the NTP Servers panel.

Then, click Disable Server. 4. To remove an NTP server, click the check box next server name in the NTP Servers panel.

Then click Remove Selected Server. To add an NTP server, do the following: 1. Select


2. Go to the Add New NTP Server panel and enter the new

NTP Server IP address in the text

box. 3. Select a Version from the drop-down list:

•

3—Network Time Protocol (Version 3)

•


Note: Ask your network administrator, if you are unsure which version to choose.

4. Specify whether or not you want the new NTP server enabled or disabled by choosing

Yes or

No from the drop-down list. 5. Click

Add NTP Server to add the server to the network.

Viewing and Modifying Web Admin Settings You can change Web Interface configurations, disable the Violin Web Interface, configure HTTP and HTTPS, set the inactivity time for automatic logouts, configure cookies, and show the Violin Web Interface configuration settings.

Caution: Disabling the Violin Web Interface terminates your Web session, after which you can only access the cluster through the command line interface. For information of the CLI, see APPENDIX A, “CLI Reference” on page 287.

126


535-0008-00 Rev 10

Administration

To disable Web management, do the following: 1. Select

Administration > Network > Web Admin. The Network Settings page appears.

Note: Web Management must be enabled by default, or you wouldn’t be able to access the

Violin cluster through the Web Interface.

2. To disable Web Management, click the Enable Web Management check box to deselect this

feature. The check mark disappears, turning off the function. 3. Click

OK to disable Web Management.

Note: After you disable the Violin Web Interface, the only way to access the Web Interface again

is to use the web enable command in the CLI. For more information, see APPENDIX A, “CLI Reference” on page 287.

To set Web configurations, do the following: 1. Select


2. Go to the Web Configurations panel, and specify the following options:

•

Enable HTTP—Enables (or disables) HTTP on the specified port.

•

HTTP Port—Port 80 is the default. Change this to a different free and unused port to access the Violin Web Interface in the following format: http:// /

•

Enable HTTPS—Enables (or disables) HTTPS on the specified port.

•

HTTPS Port—The default is 443. Change it to another available port as described in HTTP Port setting above.

•

Web UI Inactivity Timeout (secs)—Enables automatic logout and disconnect of the Web Interface connection to the cluster after a specified period of inactivity. The default auto logout timeout is 9000 seconds (15 minutes).

•

Web Session Renewal (secs)—Time in seconds before the Violin Web Interface session cookie expires.

•

Web Session Timeout (secs)—Time in seconds before the Violin Web Interface session cookie times out.

535-0008-00 Rev 10


127


3. Click

Apply to save the configurations, or Cancel to return to the previous settings.

Notification Web Interface notifications provide a means of delivering messages to a set of recipients when an event or alert occurs. For example, an e-mail can be sent to the administrator when the chassis internal temperature exceeds a specified threshold. The Notification function allows you to manage the following:

• • • • •

Call Home Email Settings Email Events SNMP Settings SNMP Events

Specifying Call Home Settings The Email (Call Home) Settings page allows you to enable email alerts in the following ways:

•

Auto Support—Enables the cluster to report alerts and critical events to a specified email address. Auto Support requires an SMTP server and email domain name.

•

Fault Reporting—Specifies the SMTP server and mail domain name used for call-home messages. These settings are also used for alerts.

To enable auto support and fault reporting, do the following: 1. Select

Administration > Notification > Call Home.

2. Select the Enable Auto Support Notifications check box. A check mark appears. To turn this

feature off, select the check box again. The check mark disappears. 3. In the Fault Reporting section, it is recommended that you keep the default

callhome.vmem.com. This ensures that Violin Memory Customer Support is notified in the case of a system critical issue. 4. Enter the Mail Domain Name in the text field, and then click

128


Apply.

535-0008-00 Rev 10

Administration

Managing E-mail Settings You can configure email settings and addresses to receive alerts from the vSHARE cluster. Call Home settings are used to send the alerts. For information on how to enable Call Home settings, see Specifying Call Home Settings on page 128. This section demonstrates the following:

• • • •

Enabling Email Alerts on page 129 Configuring the Email Server on page 130 Configuring SMTP Authentication on page 130 Managing Alert Recipients on page 131

Enabling Email Alerts

You can specify whether or not to receive alerts from the vSHARE cluster. Call Home Settings entries are used to send the alerts. For information on how to enable Call Home settings, see Specifying Call Home Settings on page 128. To enable email alerts, do the following: 1. Complete the task as described in To enable auto support and fault reporting, do the following:

on page 128. 2. Select

Administration > Notification > Email Settings.

3. In the top section of the Email Settings page, select the

535-0008-00 Rev 10


Enable Email Alerts check box.

129


4. Click

Apply to accept, or Cancel to deactivate the setting.

5. Go on to complete Configuring SMTP Authentication on page 130. Configuring the Email Server

You can configure the email server that handles event notifications in the

Note: Go to Managing E-mail Notifications on page 132 for information on how to specify the

types of events for which you want to be notified.

To configure the email server for notifications, do the following: 1. Select


2. In the Email Configuration panel, specify the following information:

• • •

Mail Hub: Enter the mail hub fully qualified domain name (FQDN), such as callhome.example.com. Mail Hub Port: Enter the port number for the mail hub.

Mail Return Address: Enter an email return address. Use do-not-reply, if you do not want to receive responses to alert notifications.

•

Mail Return Domain Name: Enter a domain name for return mail, or leave blank to use the system domain name.

•

Include Host Name to Return Address: Clear this check box to remove the host name from the return email address.

3. Click Apply to accept the

Current Return Email Address, or Cancel to modify the return

email address using the previously described fields. 4. Go on to complete Configuring SMTP Authentication on page 130. Configuring SMTP Authentication

SMTP Authentication (SMTP AUTH) is an extension of the Simple Mail Transfer Protocol (SMTP). When enabled, an SMTP client logs in using an authentication mechanism supported by the SMTP server. The following task demonstrates how to set up SMTP authentication for the e-mail server from which email alerts will be sent. To configure SMTP authentication, do the following: 1. Select 130

Administration > Notification > Email Settings. Violin 6000 Series Memory Array User’s Guide

535-0008-00 Rev 10

Administration

2. In the SMTP Authentication Configuration panel, select the

Enable SMTP Authentication

check box. A check mark appears, showing that SMTP Authentication turned on. 3. Enter the following information in the appropriate text fields:

• •

Authentication Username: The user name for the authentication credentials Authentication Password: The password for the authentication credentials

4. Click

Apply to save your configuration entries, or Cancel to delete them and start over.

5. Go on to complete Managing Alert Recipients on page 131. Managing Alert Recipients

The Alert Recipient Management section shows information about the current e-mail notification recipient, if one is configured. You can edit the account for an existing alert recipient, add new recipients, delete recipients, and send test e-mails to recipients. To edit, delete, add and verify alert recipients, do the following: 1. Select

Administration > Notification > Email Settings, and go to the Alert Recipient

Management panel.

2. To edit an existing alert recipient click

Edit next to the recipient’s name, then in the dialog that

appears, specify the following:

•

Recipient Email Address: Enter a new email address for the recipient in the text field, if needed.

•

Get Details? Select this check box to receive alerts as specified by the Call Home

•

Get Infos? Select this check box to receive less urgent informational events, or deselect

Settings page, or deselect the check box to turn off this option.

•

•

535-0008-00 Rev 10

the check box to turn off this option. Get Failures? Select this check box to receive notifications on failure events (such as file system errors, process crashes, and unexpected shutdowns), or deselect the check box to turn off this option. Click OK to activate the request, Cancel to abort the request, or Reset to clear the fields and start again.


131


3. To add a new recipient click Add a New Recipient, then in the dialog that appears, specify the

following:

•

Recipient Email Address: Enter an email address for the recipient in the text field, if needed.

•


•


•

Get Failures? Select this check box to receive notifications on failure events (such as file

Settings page, or deselect the check box to turn off this option. the check box to turn off this option.

•

system errors, process crashes, and unexpected shutdowns), or deselect the check box to turn off this option. Click OK to activate the request, Cancel to abort the request, or Reset to clear the fields and start again.

4. To delete an alert recipient click

Delete next to the recipient’s name, then in the dialog that appears click OK to confirm or No to cancel the action.

5. To send a test email to all alert recipients, click

Send Test Email and OK to approve, or

Cancel to abort the task. Managing E-mail Notifications You can individually select from the types of e-mail notification events you want to receive. The notifications are sent to specified alert recipients, as described in Managing Alert Recipients on page 131. Selecting Email Events

You can intuitively select notification events to be sent to alert recipients as specified in Managing Alert Recipients on page 131.

132


535-0008-00 Rev 10

Administration

To select events for email notification, do the following: 1. Select

Administration > Notification > Email Events.

2. Select the desired events from the list using the following actions:

•

Expand and collapse a category by clicking the arrow next to its name, or click Show All to view all events or Hide All to collapse all categories.

•

Choose from the following global event actions: Select All: Selects all events Select None: Deselects all events Reset Defaults: Returns to system default selections

•

Select individual events in a category by clicking the check box next to the name. A check mark appears when the event is selected. Select the event again to deselect the event and remove the check mark. Select all the events in a category by clicking the check mark icon on the right side of the header.

•

•

Deselect all selected events in a category by clicking the X icon on the right side of the header.

•

Reset the selections for a category to the default selections by clicking the star icon on the right side of the header.

3. Click

535-0008-00 Rev 10

Apply to accept your selections, or Cancel to return to the previous settings.


133


Managing SNMP Settings You can individually configure several types of SNMP settings through the Web Interface. Simple Network Management Protocol (SNMP) is an Internet standard protocol for managing and monitoring devices on an IP network. For more information, see APPENDIX F, “SNMP Traps” on page 433. This section covers the following:

• • •

Enabling SNMP Management on page 134 Configuring SNMP on page 134 Setting SNMP Trap Sinks on page 135

Enabling SNMP Management

You can interactively enable and disable SNMP Management from the SNMP Settings page. To enable or disable SNMP management on your system, do the following: 1. Select

Administration > Notification > SNMP Settings.

2. In the upper panel, select the Enable SNMP Management check box. A check mark appears

to show SNMP Management is turned on.

3. To disable SNMP Management, select the check box again. The check mark disappears. 4. Click

Apply to activated your selection, or Cancel to return to the previous state.

Configuring SNMP

You can configure SNMP, including settings for the Read-only Community and SNMP Traps Community.

134

•

An SNMP TRAP is a message that is sent out when a significant system event is detected. A trap sink community defines a host and a community string pair. The community string is the most basic form of SNMP security.

•

SNMP community strings are like passwords for network elements. One community string is often used for read-only access with the default set to “public.” The read-only string is used for the transfer of data between devices.

•

A read-write community string is used less often, with a default value of “private.” The readwrite community string provides the ability to change MIB variables on the device.


535-0008-00 Rev 10

Administration

To configure SNMP settings, do the following: 1. Select


2. In the SNMP Configuration panel, enter the following information the appropriate text fields:

• • •

SNMP Port: The port number assigned for SNMP traffic System Location: IP address of the SNMP server System Contact: The email address for the system administrator

3. Enable or disable the following options by selecting the check box next to the option, then from

the drop-down list choose between Public, or Private:

• •

Read-Only Community SNMP Traps Community

4. Click

Apply to activated your selection, or Cancel to return to the previous state.

Setting SNMP Trap Sinks

A trap sink defines a host and a community string pair. When a trap event occurs the SNMP agent sends out a trap to each host in the trap sink list using the community string. This section shows you how to configure a trap sink list, by removing, enabling, disabling, and adding SNMP servers (hosts) to be included in the list. The SNMP Trap Sinks panel lists configured SNMP servers, their type-version (type of security implemented as related to the version of SNMP), if SNMP management is enabled or not, and the community configuration. For more information on the available SNMP traps, see Available Traps on page 438. To remove, enable, or disable SNMP trap sink servers, do the following: 1. Select


2. To remove an SNMP server, do the following: a. Go to the SNMP Trap Sinks panel. b. Select the SNMP server from the list. c. Click

535-0008-00 Rev 10

Remove Selected Server.


135


3. To enable or disable an SNMP server, do the following: a. Go to the SNMP Trap Sinks panel. b. select the SNMP server from the list. c. Click

Enable Server or Disable Able Server.

To add new SNMP Trap Sinks, do the following: 1. Select


2. Go to the SNMP Trap Sinks panel and under Add New SNMP Trap Sinks, enter the

SNMP

Server IP address in the text field. 3. From the Type-Version drop-down list, choose one of the following:

• •

Trap-v1: Implements community-based security Trap-v2: Implements party-based security

4. From the Enable Server drop-down list, choose

Yes or No.

5. From the Community drop-down list, choose Public or Private. For more information on these

options, see Configuring SNMP on page 134.

6. Click

Add SNMP Server.

Managing SNMP Traps You can individually select from all available SNMP notification events. For information on using MIB Browsers to help manage SNMP, see Benefits of Using a MIB Browser on page 434.

Note: The notifications are sent to the email address specified on Administration > Notification >

Email Settings page. For more information, see Configuring SNMP on page 134.

Selecting SNMP Traps Events

The SNMP Traps Events page allows you to interactively select SNMP trap events for which notifications are sent. For information on the available SNMP traps, see Available Traps on page 438.

136


535-0008-00 Rev 10

Administration

To select events for email notification, do the following: 1. Select

Administration > Notification > SNMP Events.

2. Select the desired SNMP traps events using the following actions:

•

Expand and collapse a category by clicking the arrow next to its name, or click Show All to view all events or Hide All to collapse all categories.

•

Choose from the following global event actions: Select All: Selects all events Select None: Deselects all events Reset Defaults: Returns to system default selections

•

Select individual events in a category by clicking the check box next to the name. A check mark appears when the event is selected. Select the event again to deselect the event and remove the check mark. Select all the events in a category by clicking the check mark icon on the right side of the header.

•

•

Deselect all selected events in a category by clicking the X icon on the right side of the header.

•

Reset the selections for a category to the default selections by clicking the star icon on the right side of the header.

3. Click

535-0008-00 Rev 10

Apply to accept your selections, or Cancel to return to the previous settings.


137


Tools • •

Upgrade Feature Licenses

Upgrading the Violin Cluster You can upgrade the software running on the Memory Array through the Web Interface. This includes both the Memory Array firmware and the Memory Gateway software.You can choose from two upgrade scenarios: immediate reboot and staged reboot.

Note: Before attempting to upgrade the system, always review the latest Release Notes for the most up-to-date instructions, information on the necessary software image, and required upgrade paths.

To upgrade the Memory Array software, do the following: 1. Select

Administration > Tools > Upgrade.

2. Check the status of the nodes in the cluster. Resolve any problems, before continuing with the

upgrade. Note: If the node count does not match the expected number during an upgrade, you are prompted with an option to force the upgrade. Warnings of an unexpected node count also appear on the header status, cluster summary page, index page, versions page, and upgrade page.

138


535-0008-00 Rev 10

Administration

3. Do one of the following:

• •

Enter a URL or the name of an image in the Upgrade from URL or image text box. Select a file name from the Available Images drop-down list. When you select an image from this list the name of the image automatically appears in the Upgrade from URL or image text box.

Note: If you select a URL image for the upgrade and have a slow Internet connection, your login session may time out before the image download is complete. As a workaround, you can temporarily extend the timeout period on the Administration > Web Admin page. 4. Select one of the following upgrade modes by clicking the radio button:

• •

Immediate reboot: Upgrades the ACMs and the Memory Gateways immediately. This IS NOT a non-destructive upgrade. Staged reboot: Upgrades only the Memory Gateways at specified intervals, or stages. This is a non-destructive upgrade.

Note: You are logged out of your browser session when the master Violin Memory Gateway

reboots.

5. Click

Upgrade to proceed, or Abort to cancel the upgrade.

WARNING! Once an upgrade has begun, aborting the upgrade is not recommended. Doing so

may leave the cluster in an inconsistent state.

Adding and Removing Feature Licenses The Administration > Feature Licenses page allows you to add or remove feature licenses. A list of currently installed licenses appears at the top of the window. Adding a New Feature License

Adding a new feature requires a license key. You must obtain from your Violin Memory Customer Support representative before you begin the following procedure. To add a new feature license, do the following: 1. Select

Administration > Notification > Feature Licenses.

2. In the Add New Licenses panel, enter the license key for the new features in the text box. Note: If you are adding more than one license, enter each key on a new line.

535-0008-00 Rev 10


139


3. Click

Add Licenses. The new feature license appears in the Installed Licenses panel at the

top of the window. Removing a Feature License

You can easily remove installed feature licenses, either individually or all at once, using the following procedure. To remove a license, complete the following steps: 1. Select

Administration > Feature Licenses.

2. In the Installed Licenses panel at the top of the window, select the check box of the licenses

you want to remove, as shown in the following example.

3. Click

Remove Selected Licenses.

Information The Information function provides easy access to the following:

•

Documentation—Information on how to access context-sensitive online help, along with a link to a PDF version of this guide, and another link to download Adobe Acrobat Reader for viewing. This page also provides a link to the SNMP MIB files for the current software release.

•

About—Shows which version of the Memory Gateway is running and provides basic information about trademarks and patents.

• •

Violin License—Provides a complete copy of the Violin Memory license. Acknowledgements—Lists copyright and licensing information for the open source components of the product. Portions of this software product utilize open source copyrighted material; attribution to copyright holders of such materials along with applicable licensing terms are included here.

140


535-0008-00 Rev 10

Administration

Accessing Context-Sensitive Help Context-sensitive online help is available for each page in the Web Interface. To access context-sensitive Help, do the following: 1. Open the Web Interface as described in Logging In to the Web Interface on page 35. 2. On the Web Interface for which you want Help, click the

Page Help button in the upper right

corner of the window.

The help for that page appears, as shown in the following example.

3. Click

Hide Page Help to close the Page Help panel, as shown in the previous example.

Accessing a Full PDF Version of this Guide You can open a searchable PDF version of this guide from the Web Interface. Prerequisite

To view the searchable PDF version of this guide, you must have Adobe Acrobat Reader installed. You can download Adobe Acrobat Reader free from the Adobe Reader Web page. To access a full PDF version of this guide, do the following: 1. Select

Administration > Information > Documentation.

2. Click the book icon, as shown in the following example.

535-0008-00 Rev 10


141


A new tab opens in the browser, displaying a full PDF version of this guide. 3. To save a copy of the guide on your local system, click the Save icon in the upper left corner of

the toolbar, navigate to the appropriate location, and click Save.

Viewing System, License, and Copyright Information You can easily access information on the system you are logged in to, your Violin Memory license, as well as all copyrights and other legal notices. To view system, license, and copyright information, do the following: 1. To view system information, select

Administration > Information > About. A page appears

showing the current version of vSHARE running on your system. 2. To view information on your Violin Memory End User License, select

Administration >

Information > Violin License. 3. To view applicable legal notices and acknowledgements for this release, select

Administration > Information > Acknowledgements.

142


535-0008-00 Rev 10

CHAPTER 3

Manage the Memory Array with the Web Interface

This chapter provides basic Memory Array monitoring and administrative tasks in the following sections, focusing on completing the tasks using the Violin Web interface. If CLI commands are available for a task, references to those sections are provided.

• • • • • • •

Introduction to the Memory Array Web Interface on page 143 Logging In to and Out of the Array Web Interface on page 146 Performing Tasks Using the Web Interface on page 148 Viewing Status of the Memory Array on page 148 Monitoring the Memory Array on page 152 Administration on page 154 Monitoring the Memory Array Using the XTree Browser Interface on page 173

See Manage, Monitor, and Administer with the Violin Web Interface on page 31 for instructions on managing the Memory Gateways that are internal to the Memory Array.

Introduction to the Memory Array Web Interface The Memory Array interface is a web-based interface that you can access by connecting to the Master Array Controller Module (ACM) of the cluster using a Web browser. The screens available to you depend on your user privileges. See User Roles on page 158 for an overview of the privileges available to the two user roles.

535-0008-00 Rev 10


143

Chapter 3: Manage the Memory Array with the Web Interface

High-Level Overview The streamlined Web interface includes the following interactive components, and is shown in Figure 3.1.

•

Main Menu Bar—Displays clickable icons for the major functional categories: Home, Monitor and Administration.

•

Function Menu Bar—Changes to display the options for the currently active category. Click an icon to display the associated page and submenu options.

• • •

Online Help—Brings up context-sensitive help for the currently active page.

•

Software Version—Displays the version of the software currently running on the Memory Array.

Session—Shows the role of the current user at the right, and exits the session when clicked. Board Status Tab—Expands to show status of all the Memory Array modules when clicked. Click the tab again to hide the Board Status page.

Main Menu Bar

Function Menu Bar

Online Help

Session

Board Status Tab

Software Version

Figure 3.1 Web Interface Organization

144


535-0008-00 Rev 10

Introduction to the Memory Array Web Interface

Web Interface Requirements This section provides information on the Violin Web interface requirements for browsers, display resolution, Adobe Flash Player, JavaScript and cookies. System and Browser Requirements The following table lists the supported browsers and versions for the Violin Web interface:

Operating System

Linux Mac

Windows

Supported Browser

• •

Mozilla Firefox 4 and above

• • •

Apple Safari 4 and above

• • • •

Windows Internet Explorer 9 and above

Google Chrome 11 and above Mozilla Firefox 4 and above Google Chrome 11 and above Apple Safari 4 and above Mozilla Firefox 4 and above Google Chrome 11 and above


The following table lists the recommended software and settings for optimum performance of the Violin Web interface.

Recommendation

Supported Browser

Display resolution

The minimum recommended display resolution for the Violin Web interface is 1024 x 768 pixels

JavaScript

JavaScript must be enabled for the Violin Web interface

Adobe Flash Player

Adobe Flash Player version 8 (or above) is required for viewing charts in the Violin Web interface

Cookies

Enable cookies for login and session management


Web Interface Help There are two types of available help:

•

Context-sensitive online help for any page in the Web interface—Click the Page Help button in the upper right corner of the Web interface.

•

A full PDF version of this guide—Select Administration > Information > Documentation and then click the guide cover image to view the PDF.

For more information on available Help, see Information on page 171.

535-0008-00 Rev 10


145


Configuring the Violin Web Interface Configuration of the Violin Web Interface can be changed by using the Administration > Network > Web Admin screen or the “web” commands in the CLI. You can enable or disable the Violin Web Interface, configure HTTP and HTTPS, set the inactivity time so that you are automatically logged out when the interval is reached, configure cookies, and show the Web Interface configuration settings. See Viewing and Modifying Web Admin Settings on page 167 or APPENDIX A, “CLI Reference” on page 287 for descriptions of the CLI commands.

Printing Screens You can print any screen in the Violin Web Interface by typing CTRL+P and then selecting Print from the menu.

Logging In to and Out of the Array Web Interface You must enter a valid user name and password on the Login screen before you can view any other screens. These credentials are set by the administrator. Contact your administrator if you do not have a user name and password assigned. To access the Violin Web Interface and log in, do the following: 1. Open a Web browser, as described in System and Browser Requirements on page 145. 2. In the browser URL field, enter one of the following:

•

http://

•

http://

If you specify the IP address or host name of a cluster node (other than the master), the connection is automatically redirected to the Master ACM (Array Controller Module). The Web Interface log in page appears.

146


535-0008-00 Rev 10

Logging In to and Out of the Array Web Interface

3. Enter a valid user name and password in the text fields of the Login page, as shown in the

following example.

4. Click Login. The Array Status page appears. 5. Continue with Viewing Status of the Memory Array on page 148.

Logging Out of the Web Interface You can log out of the Array from any page in the Web interface. To log out of your Web Interface Memory Array session, do the following: 1. From any page in the Web interface, go to the upper right hand corner. 2. Click the session icon, as shown in the following example.

The Logout page appears acknowledging that you successfully logged out.

535-0008-00 Rev 10


147


To log back in, click Login Again and re-enter your credentials.

Performing Tasks Using the Web Interface Some of the tasks described in this chapter can be performed using the Web interface and/or the Command Line Interface (CLI). The table below lists those tasks and where to find additional information about both. Task

Web UI

CLI

Viewing Status of the Memory Array

page 148

page 321

Getting Help

page 172

page 310

Rebooting the Memory Array

page 154

page 239

Powering Off and On Memory Array Modules

N/A

page 240

Table 3.3 Task Map

Viewing Status of the Memory Array The Array Status screen appears when you log in and when you click Home. The screen contains a top section for information about the system as a whole and a lower section—displayed by clicking Show Board Status tab—that provides information about every module that is present on the main system board. See Showing Status of the Memory Array Modules on page 149. This section can be minimized by clicking the Hide Board Status button.

148


535-0008-00 Rev 10


Figure 3.2 Array Status Page

The Status screen provides the following status information:

•

The system name and ID, with the total number of VIMMs in the cluster.

The overall status of the system, includes:

• • • • •

how many VIMMs are in various states (active, spare, booting, etc.) temperatures of ambient air, controller, and VIMMs (for each Array) fan status indicators estimated lifetime of flash storage status lights: Alarm, Status, and Ready

To view Array statistics, do the following: 1. Select

Home > Array Status. The Array Status page appears.

2. Check alarm status at the right side of the panel. 3. Verify the life span of the flash in the Flash Lifetime panel. 4. Check the health of the Fans. A failed fan displays as red. A problematic fan displays as

yellow. A green box with a white check mark represents a healthy, fully functioning fan. 5. Check the VIMM, Controller, and ambient temperatures. Green is healthy, yellow a caution,

and red a warning. 6. View a high-level synopsis of the flash VIMM statistics in the far left panel.

Showing Status of the Memory Array Modules You can display a top-down view of all Memory Array modules from the Array Status page. 1. Select

Home > Array Status. The Array Status page appears.

1. Click the Board Status tab.

535-0008-00 Rev 10


149


All of the modules are color-coded with indicator boxes for alarm status, temperature, and remaining lifetime.

Figure 3.3 Board Status Screen

You can obtain general information about a module by placing the cursor over the module on the Board Status screen. Click an individual module for more detailed information.

Viewing VIMM Status The Array Status page can be expanded to view board-level information about a Violin Memory Array. The Board Status page shows a color-coded array of VIMMs with indicator boxes for alarm status, temperature, and remaining lifetime.You can move your cursor over a VIMM to view a detailed status. To view VIMM statistics, do the following: 1. Select

Home > Array Status, to display the Array Status page.

2. Click the

150



535-0008-00 Rev 10


3. Place the cursor over the VIMM for general information. Click the VIMM for more detailed

information about the VIMM. An example is shown below.

Above the VIMMs, you can click a RAID Group number to focus on the VIMMs within the selected RAID group. Below the VIMMs, you can click a “Show” category to highlight the following types of VIMMs:

• • • •

failed booting admin down spares

Focusing on VIMMs Managed by vRAID Controllers The Board Status Display provides additional views that allow you to identify which VIMMs are managed by which vRAID Controller Module (VCM). 1. Select


2. Click the


3. Click the letter next to “VIMM Controller” on which you want to focus (A, B, C or D).

535-0008-00 Rev 10


151


Figure 3.4 shows the VIMMs managed by VCM A.

Figure 3.4 Focus on VIMM Controller

Focusing on Modules Managed by Power Controllers The Board Status Display provides additional views that allow you to identify which VIMMs are managed by which Power Controllers. 1. Select


2. Click the


3. Click the Power Controller letter (A or B) on the bottom left of the Board Status screen.

Monitoring the Memory Array The Monitor function provides an at-a-glance view of the health and performance of the Memory Array, and the ability to analyze historical data and real time trends. The Monitoring page provides a high-level view of real-time or historical data. You can customize how you view the data, choosing between line and area graphs, time trends, as well as the type of metrics used to formulate the data.

152

•

Real Time Performance Trend—Shows the last five minutes of performance trends for the selected metric. New real time values are appended into the graph every ten seconds.

•

Current Value—Shows real time data at this time. The graph time is listed at the bottom of the panel.


535-0008-00 Rev 10

Monitoring the Memory Array

Customizing the Output Display The following task demonstrates how to customize the data display to suit your needs. You can specify the visual graph you prefer, the time trend, and the applied metrics. To customize the data output display, do the following: 1. Select

Monitor > Monitoring. The Monitoring page appears.

2. To specify the real-time performance trend, at the top of the left-most panel choose one of the

following options:

•

Last 1 H: Shows data averaged every one minute.

•

Last 24 H: Shows data averaged over the last 24 hours.

3. To view detailed information for a point in time, move the cursor over a point on the graph. A

pop-up dialog appears, as shown in the following example.

Saving a Snapshot of Performance Data 1. Select

Monitor > Monitoring. The Monitoring page appears.

2. To save a snapshot of performance data, do the following: a. Click the

Download button. b. Choose a Time Interval from the drop-down list. c. Click Download as CSV. Data for the specified time interval is saved as a CSV (comma separated value) file. Once saved, the file can be opened using Microsoft Excel.

535-0008-00 Rev 10


153


Administration The Administration page of the Violin Web Interface provides a single point of contact where you can perform a wide range of Memory Array administration tasks. You must have Admin privileges to perform the following tasks using the Administration function pages. Function Page

Array Admin

Tasks

• • • • • •

Rebooting and Shutting Down the Memory Array on page 154

• • • • • •


• • •


Tools

•


Information

• • • •


Network

Notification

Changing and Saving Configuration Files on page 155 Viewing ACMs on page 156 Viewing Alerts on page 156 Searching for and Viewing Logs on page 157 Viewing ACM Versions on page 157 Changing the Cluster Management Address on page 163 Modifying Global Default Gateway Settings on page 164 Viewing and Modifying DNS Settings on page 164 Viewing and Modifying NTP Settings on page 165 Viewing and Modifying Web Admin Settings on page 167 Managing User Accounts on page 158 Managing E-mail Settings on page 169

Accessing a Full PDF Version of this Guide on page 172 Viewing System, License, and Copyright Information on page 173 Monitoring the Memory Array Using the XTree Browser Interface on page 173

Array Admin Array administration includes configuration and management tasks for the Memory Array. Rebooting and Shutting Down the Memory Array The Cluster Administration page displays information on the current software version, the installed software image, and configuration files. Two boot images are installed for each ACM in separate partitions. The Software section displays the software version number and allows you to choose which boot image the ACMs should use (the Switch Boot Partition button). Two boot images are installed for each ACM in separate partitions.

154


535-0008-00 Rev 10

Administration

Note: Rebooting the system will take several minutes to complete. 1. Select

Administration > Array Admin > Reboot. The Cluster Administration page appears.


— To reboot the system, click Reboot. — To shut down the system, click Shutdown. A pop-up dialog appears. 3. Click

OK to continue with the reboot process, or Cancel to quit without rebooting or shutting

down the system. See Reboot and Shutdown Procedures on page 238 for instructions on performing these operations using the CLI.

Changing and Saving Configuration Files You can save the current Memory Array configuration to a specified file, apply a previously saved configuration to the Memory Array (whether a local file or on a remote server), and delete old configuration files. The currently active configuration is marked as Active in the configuration table. The Status column indicates which configuration file is active. The active file is backed up as .bak. To save, apply, or delete a Memory Array configuration file, do the following: 1. Select

Administration > Array Admin.

2. To apply a previous configuration to the Memory Array, select the check box next to the Config

Name and then click Apply.

3. To save the current Memory Array configuration, enter a unique name in the

Save

configuration as field, and then click Save.

To delete a configuration, select the check box next to the Config Name and click Delete.

535-0008-00 Rev 10


155


Viewing ACMs View ACMs to see a list of all nodes in the cluster. The status of each ACM is normally online although transition states also occur while the ACM is joining or leaving the cluster.

•

Select Administration > Array Admin > ACMs. The ACM List page appears.

The Master ACM is the current master control node in the cluster. It works together with a Standby to manage the availability of the cluster. To view more detail about an ACM:

•

Click View to show more details about an individual ACM. Alternatively, you can show the details by clicking the ACM ID in the ACM column.

This screen shows the software version running on the ACM, its IP addresses, status and role. It also shows how long the ACM has been running (Uptime).

Viewing Alerts The Alerts page shows a listing of current and past system events that triggered an alert. Select the types of alerts to display (informational, warning, or error) and click the APPLY button. To view alerts, do the following: 1. Select

156

Administration > Array Admin > Alerts.


535-0008-00 Rev 10

Administration

2. To change the types of alerts that appear in the list, deselect or select the check box next to

the Show icon: Info, Warning, Error. A check mark indicates that the alert type is selected to appear in the list. 3. Click the

Next to page forward through the list and Prev to go back.

4. To change the number of alerts displayed on a page, choose a number from the Display drop-

down list and then click Apply. 5. To sort based on the values in a column, click the column heading. Click the heading a second

time to reverse the sorting order.

Searching for and Viewing Logs The Web interface allows you to easily view a listing of logged events. You can choose between viewing current and historical logs in paginated format. Log files are rotated once a day. To view logs, do the following: 1. Select

Administration > Array Admin > Logs. The Memory Array Logs page appears.

2. To navigate to the desired log page, do any of the following:

— Click Next (>) to page forward through the log files and Prev ( Array Admin > Versions. The Versions page appears with a list of

ACMs and the following information for each:

535-0008-00 Rev 10


157


— — — — —

ACM ID: ACM name Uptime: How long the ACM has been running Version: Software version running on the ACM Boot/Next: The boot ACM and next ACM to be booted Array ID String: Current boot image

To view more details on a ACM, click the ACM ID link on the left. The View ACM page appears, as described in Viewing ACMs on page 156.

Managing User Accounts User Management displays the name, role, email address, and password status of each user account for the Memory Array. This section demonstrates how to perform the following tasks:

• • • •

Adding a User and Specifying a Role on page 159 Changing an Account Password on page 160 Editing a User Account on page 160 Deleting a User Account on page 161

Understanding User Roles and Privileges

The system administrator can permit others to access the Memory Array Web interface and the CLI. This section provides an overview of the two user roles and the privileges available to each. User Roles The Memory Array has two built-in system accounts:

• •

admin with full privileges. monitor with privileges to read data and perform limited actions.

Caution: Initially these accounts have no password. Setting a password for “admin” is highly

recommended.

A user’s assigned role determines the level of privileges they have in accessing and administering the Memory Array. Some of parameters are only visible or configurable by users with certain roles

158


535-0008-00 Rev 10

Administration

(privileges). The available roles and their privileges are as follows. Privileges are set by assigning one of the following roles to a user account: admin:

The administrator role has full privileges to view anything, take any action, or change any configuration. This role can access every screen in the Violin Web Interface.

monitor:

The monitor role can read all data and perform some actions, such as rebooting the system and configuring various system parameters, but cannot change the configuration of the Violin Memory Array. This role can view some of the screens in the Web Interface.

The CLI command modes Config, Enable, and Standard correspond to the user roles admin, monitor, and unpriv, respectively. You can add user accounts and set their privileges and passwords in the Administration section of the Violin Web Interface. Adding a User and Specifying a Role

An Admin user can add new users and define their account privileges. To add a new user account and define the access privileges, do the following: 1. Select

Administration > Notification > Users.

2. At the bottom of the page, click Add

New User. The Add New User dialog appears.

3. Specify the following account information:

• •

User Name: Enter a user name for login credentials. User Role: Select a role from the drop-down list, admin or monitor. For more information, see Understanding User Roles and Privileges on page 158.

4. Click

OK to add the new account, Cancel to abort the action, or Reset to clear the fields and start over.

535-0008-00 Rev 10


159


Changing an Account Password

An Admin user can set or reset the passwords and define access privileges for other user accounts. To add a new user account and define the access privileges, do the following: 1. Select


2. Next to the user name, click

Change Password. The Change Password dialog appears.

3. Specify the following information:

• • •

User Name: Enter the user name for the login credential. Password: Enter the new password. Confirm: Re-enter the new password.

4. Click

OK to accept the change, Cancel to abort the action, or Reset to clear the fields and start over.

Editing a User Account

An Admin user can change an existing user account, modifying personal and contact information. To edit an existing user account, do the following: 1. Select 2. Click


Edit for the desired user. The Edit Existing User dialog appears.

3. Change the following account information, as necessary:

• • • •

160

Role: Select a role from the drop-down list, admin or monitor. For more information, see Understanding User Roles and Privileges on page 158. First Name: Enter the user’s first name.

Last Name: Enter the user’s last name. Email Address: Enter the user’s email address.


535-0008-00 Rev 10

Administration

4. Click

OK to accept the changes, Cancel to abort the action, or Reset to clear the fields and start over. If you accept the changes, another dialog appears.

5. Click

OK in the confirmation dialog to save the changes. The results appear in the User Management panel.

Deleting a User Account

An Admin user can delete the accounts of other users. To delete a user account, do the following: 1. Select 2. Click


Delete next to the account name you want to remove. A confirmation dialog appears.

.

3. Click

Yes to confirm your choice, or No to cancel the action. If you click Yes, the user account is deleted from the list.

535-0008-00 Rev 10


161


Managing E-mail Settings You can configure e-mail settings and addresses to receive alerts from the Memory Array. Call Home settings are used to send the alerts. For information on how to enable Call Home settings, see Specifying Call Home Settings on page 168. The Alert Recipient Management section shows information about the current e-mail notification recipient, if one is configured. You can edit the account for an existing alert recipient, add new recipients, delete recipients, and send test e-mails to recipients. To edit, delete, add and verify alert recipients, do the following: 1. Select





•

Recipient E-mail Address: Enter a new email address for the recipient in the text field, if needed.

• • •

•

Get Details? Select this check box to receive alerts as specified by the Call Home Settings page, or deselect the check box to turn off this option. Get Infos? Select this check box to receive less urgent informational events, or deselect the check box to turn off this option. Get Failures? Select this check box to receive notifications on failure events (such as file system errors, process crashes, and unexpected shutdowns), or deselect the check box to turn off this option. Click OK to activate the request, Cancel to abort the request, or Reset to clear the fields and start again.


following:

•

Recipient Email Address: Enter an email address for the recipient in the text field, if

•


•

Settings page, or deselect the check box to turn off this option. Get Infos? Select this check box to receive less urgent informational events, or deselect the check box to turn off this option.

needed.

162


535-0008-00 Rev 10

Administration

•

•

Get Failures? Select this check box to receive notifications on failure events (such as file system errors, process crashes, and unexpected shutdowns), or deselect the check box to turn off this option. Click OK to activate the request, Cancel to abort the request, or Reset to clear the fields and start again.

To delete an alert recipient click Delete next to the recipient’s name, then in the dialog that appears click Yes to confirm or No to cancel the action.

Network The Network Web interface pages provide network configuration information for the Memory Array, and allows you to manage the cluster address and global default gateway. The tasks you can perform on these pages are intended for network administrators, and require Admin user access privileges. The cluster management address is the IP address for the Violin Memory Array cluster.

Viewing Current Network Settings The Network Settings page allows you to view and modify network settings from one central location. To view current cluster network settings, do the following: 1. Select

Administration > Network > Network Settings. The Network Settings page appears.

2. Verify the

Cluster Management Address and Network Mask Length.

The IP address for cluster management is set as part of the initial out-of-the box configuration. You can change the cluster management IP address and network mask length in this section, but it is not recommended. Consult with your network architect before changing these values. 3. Review the

Bonding Mode setting.

This refers to Cluster Interface Bonding. 4. Review the

Routing settings.

This is the address for the global default gateway. You can enable and disable the gateway and set a new address for the default gateway. 5. Review the cluster

Static and Dynamic Routes.

This is the routing information for the Memory Array, including destination, mask, gateway, whether the node is active or static, and the interface.

Changing the Cluster Management Address Caution: Changing the cluster management address is not recommended. Consult with your

network administrator before changing the cluster management address. If you change this address through the Web Interface, you lose connectivity with the cluster.

535-0008-00 Rev 10


163


You can change the cluster management address from the Network Settings page. In most cases, it is recommended that you change the cluster management address through a serial console using the CLI. However, there are few situations where changing the cluster management address through the Web Interface is advised, one being in preparation for a physical move of the devices. To change the cluster management address, do the following: 1. Select


2. In the Cluster Management Address panel, enter an 3. Enter the 4. Click

IP Address in the text field.

Network Mask Length in the text field.

Apply.

Modifying Global Default Gateway Settings The Network Settings page allow you to specify the Global Default Gateway and then enable or disable its use. To modify the global default gateway, do the following: 1. Select


2. To disable the current global default gateway, click

Disable. To re-enable the default gateway,

click Enable. 3. To change the global default gateway, enter the new default gateway address in the text field

and click Set Default Gateway.

Viewing and Modifying DNS Settings The DNS Servers screen shows the list of DNS servers available to the Memory Array. You can view domain name server (DNS) configurations, add DNS servers, as well as add and remove domain names from the DNS Settings page.

Note: If one or more IP addresses for DNS servers were set up as part of the initial out-of-the-

box setup, they are listed on this page.

To view configured DNS settings, do the following: 1. Select


2. Go to the DNS Servers panel at the top of the window to view information on the DNS servers

that were configured as part of the initial out-of-the-box setup.

164


535-0008-00 Rev 10

Administration

To add and modify name servers, do the following: 1. Select


2. To modify a configured name server, go to the Add or Modify Name Servers panel and change

the Primary DNS IP address displayed in the text box. 3. To add a second name server, enter an IP address in the

Secondary DNS IP address text

box. 4. To add a third name server, enter an IP address in the

5. Click

Tertiary DNS IP address text box.

Apply.

To add and remove domain names, do the following: 1. Select


2. To remove an existing domain name, go to the

Domain Names panel and select the domain

name from the list. The selected name highlights.

3. Click

Remove Selected.

Viewing and Modifying NTP Settings The Network Time Protocol (NTP) is a networking protocol for synchronizing the clocks of computer systems over data networks. You can enable or disable NTP synchronization and add or remove an NTP server through the NTP Settings page. When you initially go to the NTP Settings page, information for the NTP server that was configured in the out-of-box setup is shown. To enable or disable NTP time synchronization, do the following: 1. Select


2. To enable NTP, click the

Enable NTP Time Synchronization check box. A check mark

appears to show it is enabled. Then click Apply.

535-0008-00 Rev 10


165


3. To disable NTP time synchronization, click the check marked

Enable NTP Time Synchronization check box. The check mark disappears to show that it is disabled. Then click Apply.

To enable, disable, or remove an NTP server, do the following: 1. Select


2. To enable an NTP server, click the check box next to the server name in the NTP Servers

panel. Then, click Enable Server.

3. To disable an NTP server, click the check box next to the server name in the NTP Servers

panel. Then, click Disable Server. 4. To remove an NTP server, click the check box next to the server name in the NTP Servers

panel. Then click Remove Selected Server. To add an NTP server, do the following: 1. Select


2. Go to the Add New NTP Server panel and enter the new

NTP Server IP address in the text

box. 3. Select a Version from the drop-down list:

•


•


Note: Ask your network administrator, if you are unsure which version to choose.

166


535-0008-00 Rev 10

Administration

4. Specify whether or not you want the new NTP server enabled or disabled by choosing

Yes or

No from the drop-down list. 5. Click

Add NTP Server to add the server to the network.

Viewing and Modifying Web Admin Settings You can change Web interface configurations, disable the Violin Web Interface, configure HTTP and HTTPS, set the inactivity time for automatic logouts, configure cookies, and show the Violin Web Interface configuration settings.

Caution: Disabling the Violin Web interface terminates your Web session, after which you can only access the Memory Array through the command line interface. For information of the CLI, see APPENDIX A, “CLI Reference” on page 287.

To disable Web management, do the following: 1. Select


Note: Web Management must be enabled by default, or you wouldn’t be able to access the

Memory Array through the Web interface.

2. To disable Web Management, click the Enable Web Management check box to deselect this

feature. The check mark disappears, turning off the function. 3. Click

Apply to disable Web Management.

Note: After you disable the Web management, the only way to access the Web Interface again is to use the web enable command in the CLI. For more information, see APPENDIX A, “CLI Reference” on page 287.

To set Web configurations, do the following: 1. Select


2. On the Web Configurations panel, specify the following options:

•

Enable HTTP—Enables (or disables) HTTP on the specified port. Port 80 is the default. Change this to a different free and unused port to access the Violin Web Interface in the following format: http:// /

•

Enable HTTPS—Enables (or disables) HTTPS on the specified port. The default is 443. Change it to another available port as described in HTTP Port setting above.

535-0008-00 Rev 10


167


•

Web UI Inactivity Timeout—Enables automatic logout and disconnect of the Web interface connection to the Memory Array after a specified period of inactivity. The default auto logout timeout is 900 seconds (15 minutes).

•

Session Renewal—Time in seconds before the Web interface session cookie expires.

•

Session Timeout—Time in seconds before the Web interface session cookie times out.

Click Apply to save the configurations, or Cancel to return to the previous settings.

Notification Web interface notifications provide a means of delivering messages to a set of recipients, when an event or alert occurs. For example, an e-mail can be sent to the administrator when the chassis internal temperature exceeds a specified threshold. The Notification function allows you to manage the following:

• •

Call Home Email Settings

Specifying Call Home Settings The Email (Call Home) Settings page allows you to enable email alerts in the following ways:

•

Auto Support—Enables the Memory Array to report alerts and critical events to a specified email address. Auto Support requires an SMTP server and email domain name.

•

Fault Reporting—Specifies the SMTP server and mail domain name used for call-home messages. These settings are also used for alerts.

To enable auto support and fault reporting, do the following: 1. Select

Administration > Notification > Call Home.

2. Select the Enable Auto Support Notifications check box. A check mark appears. To turn this

feature off, select the check box again. The check mark disappears. 3. In the Fault Reporting section, enter an SMTP Server address in the text field, such as

callhome.example.com. 4. Enter the Mail Domain Name in the text field.

168


535-0008-00 Rev 10

Administration

5. Click

Apply.

Managing E-mail Settings You can configure e-mail settings and addresses to receive alerts from the Memory Array. Call Home settings are used to send the alerts. For information on how to enable Call Home settings, see Specifying Call Home Settings on page 168. The Alert Recipient Management section shows information about the current e-mail notification recipient, if one is configured. You can edit the account for an existing alert recipient, add new recipients, delete recipients, and send test e-mails to recipients. To edit, delete, add and verify alert recipients, do the following: 1. Select





•

Recipient E-mail Address: Enter a new email address for the recipient in the text field, if

•


needed.

• •

•

535-0008-00 Rev 10

Settings page, or deselect the check box to turn off this option. Get Infos? Select this check box to receive less urgent informational events, or deselect the check box to turn off this option. Get Failures? Select this check box to receive notifications on failure events (such as file system errors, process crashes, and unexpected shutdowns), or deselect the check box to turn off this option. Click OK to activate the request, Cancel to abort the request, or Reset to clear the fields and start again.


169



following:

•

Recipient Email Address: Enter an email address for the recipient in the text field, if needed.

•


•


•

Get Failures? Select this check box to receive notifications on failure events (such as file

Settings page, or deselect the check box to turn off this option. the check box to turn off this option.

•

system errors, process crashes, and unexpected shutdowns), or deselect the check box to turn off this option. Click OK to activate the request, Cancel to abort the request, or Reset to clear the fields and start again.

4. To delete an alert recipient click

Delete next to the recipient’s name, then in the dialog that appears click Yes to confirm or No to cancel the action.

Tools The Tools pages provide the functions to update or remove feature licenses and upgrade the Memory Array software.

Adding and Removing Feature Licenses The Administration > Feature Licenses page allows you to add or remove feature licenses. Some features require that a license key be present for use. A list of currently installed licenses appears at the top of the window. Adding a New Feature License

Adding a new feature requires a license key. You must obtain a license key from your Violin Memory Customer Support representative before you begin the following procedure. 1. Select

Administration > Tools > Feature Licenses.

2. In the Add New License(s) panel, enter the license key for the new features in the text box. Note: If you are adding more than one license, enter each key on a new line.

170


535-0008-00 Rev 10

Administration

3. Click

Add Licenses. The new feature license appears in the Installed Licenses panel at the

top of the window. Removing a Feature License

You can easily remove installed feature licenses, either individually or all at once, using the following procedure. 1. Select

Administration > Tools > Feature Licenses.

2. In the Installed Licenses panel at the top of the window, select the check box of the licenses

you want to remove, as shown in the following example.

3. Click

Remove Selected Licenses.

Information The Information function provides easy access to the following:

•

Documentation—Information on how to access context-sensitive online help, along with a link to a PDF version of this guide, and another link to download Adobe Acrobat Reader for viewing. This page also provides a link to the SNMP MIB files for the current software release. See About Violin Memory SNMP MIBs on page 433 for more information.

•

About—Shows which version of the Memory Array software is running and provides basic information about trademarks and patents.

• •

Violin License—Provides a complete copy of the Violin Memory license. Acknowledgements—Lists copyright and licensing information for the open source components of the product. Portions of this software product utilize open source copyrighted material; attribution to copyright holders of such materials along with applicable licensing terms are included here.

535-0008-00 Rev 10


171


Accessing Context-Sensitive Help The Memory Array provides context-sensitive online help for each page in the Web interface. 1. Open the Web Interface as described in Logging In to and Out of the Array Web Interface on

page 146. 2. On the Web interface page for which you want Help, click the

Page Help button in the upper

right corner of the window.

The help for that page appears, as shown in the following example.

3. Click

Hide Help to close the Page Help panel, as shown in the previous example.

Accessing a Full PDF Version of this Guide You can open a searchable PDF version of this from the Web interface. Prerequisite

To view the searchable PDF version of this guide, you must have Adobe Acrobat Reader installed. You can download Adobe Acrobat Reader free from the Adobe Reader Web page. To access a full PDF version of this guide, do the following: 1. Select

Administration > Information > Documentation.

2. Click the book icon, as shown in the following example.

172


535-0008-00 Rev 10

Monitoring the Memory Array Using the XTree Browser Interface

A new tab opens in the browser, displaying a full PDF version of this guide. 3. To save a copy of the guide on your local system, click the Save icon in the upper left corner of

the toolbar, navigate to the appropriate location, and click Save.

Viewing System, License, and Copyright Information You can easily access information on the system you are logged in to, your Violin Memory license, as well as all copyrights and other legal notices. To view system, license, and copyright information, do the following: 1. To view system information, select

Administration > Information > About. A page appears showing the current Memory Array software version running on the system.

2. To view information on your Violin Memory End User License, select

Administration >

Information > Violin License. 3. To view applicable legal notices and acknowledgements for this release, select

Administration > Information > Acknowledgements.

Monitoring the Memory Array Using the XTree Browser Interface An XTree browser interface is available as an alternative method for monitoring the Memory Array. This interface provides links that display information about Violin Memory Gateways and Memory Arrays, similar to issuing commands from the CLI. To log in to the XTree interface: 1. Log in to the Violin Web interface. See Logging In to and Out of the Array Web Interface on

page 146. 2. Once logged in to the Memory Array, place your pointer in the address bar at the end of the

Memory Array host name, and then replace what is to the right of the host name with the following: /xtree/*

535-0008-00 Rev 10


173


For example:

You should see the Status page. 3. Place your pointer in the address bar in front of the asterisk and then replace what is to the

right with the following: /array/** For example:

You are now logged into the XTree interface. From here you can view values for and gain information about many of the Memory Array components, and click a path name to narrow your search. To return to the Violin Web Interface, remove what is to the right of the host name from the address bar and then press Return.

174


535-0008-00 Rev 10

CHAPTER 4

vSHARE Configuration

This chapter describes vSHARE, its features and functions, and the steps required to configure a Memory Gateway running vSHARE to manage block storage. The following topics are covered:

• • • • • • • •

Understanding vSHARE on page 175 AIX Compatibility NACA Bit Support on page 179 Configuring Storage Containers on page 180 Configuring Target Ports on page 182 Configuring Initiator Groups on page 187 Working with LUNs on page 190 Configuring Storage Containers on page 180 Working with LUN Groups on page 197

Understanding vSHARE vSHARE is a solution for block storage management. vSHARE runs as software on a Memory Gateway enabling host systems (for example, database servers) to use the following transport protocols to access logical units of data (LUNs) stored within Memory Arrays:

• • •

iSCSI Fibre Channel (FCP) InfiniBand (SRP)

On the host system, the LUN appears as a local SCSI disk. The host formats and partitions the LUN. The storage system sees the contents of a LUN as a set of blocks of arbitrary data. Administrators may use vSHARE to define rules for controlling access to LUNs based on the containers, initiators and initiator groups, and target ports configured within the system. 535-0008-00 Rev 10


175

Chapter 4 - vSHARE Configuration

•

Storage Containers: A storage container comprises a Violin Memory Array. Every LUN created and managed by vSHARE is created within a storage container. You can create many LUNS within a container. However, currently there is a one-to-one (1:1) mapping between an array and a container (one container per array).

•

Initiators and Initiator groups: An initiator is a host or client that requires access to the LUNs stored in the containers. Multiple initiators may be grouped together in an initiator group (igroup). Access to the LUNs in a container may be restricted to specific initiators or initiator groups.

•

Targets: A target identifies the port on the Memory Gateway through which the initiator may access the LUNs stored in a container. Access to the LUNs in a container may be restricted by target port.

vSHARE System Architecture The vSHARE architecture is ideal for applications that require rapid access to and processing of large amounts of data including enterprise databases, data warehouses, application servers, and Cloud-scale analytics. A typical vSHARE cluster consists of one or more Memory Gateways running the vSHARE software application and one or more Violin Memory Arrays.

Figure 4.1 vSHARE System Architecture

The vSHARE application runs on Memory Gateways and allows you to manage LUN and initiatortarget relationships. The Memory Gateways operate as the SAN targets and the hosts (such as a database server or application server) are the initiators. The Memory Gateway provides network connectivity, utilizing Fibre Channel, InfiniBand, and iSCSI protocols. The Memory Gateway provides high performance processing, high bandwidth DRAM, and high bandwidth access to the Violin Memory Arrays. All LUNs created and managed by vSHARE. are stored in containers, each container comprises a Violin Memory Array that utilizes flash memory storage. For maximum performance, Single Level Cell (SLC) flash is used. For capacity to support readcentric applications, Multiple Level Cell (MLC) should be used. The Memory Gateways and Memory Arrays are connected via PCIe.

176


535-0008-00 Rev 10

vSHARE Configuration Overview

Support for vSHARE InfiniBand InfiniBand (IB) target mode allows you to create LUNs within attached Violin Memory Arrays utilizing the SCSI RDMA Protocol (SRP). SRP allows one computer to access SCSI devices attached to another computer through remote direct memory access (RDMA) for higher throughput and lower latency.

Note: Violin vSHARE IB addresses follow this format: guid.aaaa:bbbb:cccc:dddd. InfiniBand

ports are hca-xy (where x=[a,b], y=[1,2]).

To use the SRP protocol with vSHARE InfiniBand, clients must have the SRP initiator software and networking hardware listed in this section. Hardware Quad Data Rate (QDR) InfiniBand (IB) Host Channel Adapters (HCA) and switches are supported. In particular, the following QDR IB switches are supported:

• •

Mellanox IS5000 family 36-port QSFP 40Gb/s IB switch (IS5025, IS5030 or IS5035) Qlogic 12200 Series QDR 36-port Edge Switch

Software The following software SRP initiators are supported:

• • •

VMware SRP initiator, (Mellanox OFED Drivers for VMware vSphere 4 and vSphere 5). Linux SRP initiator, available since November 2005. Windows SRP initiator, available through the WinOF InfiniBand stack.

Target Ports You can view information for InfiniBand target ports using the show targets command. For more information, see show targets Command on page 186. You can view target statistics using the show stats ib command. For more information, see iSCSI Commands on page 325.

vSHARE Configuration Overview Violin vSHARE is the software application that provides block storage functionality on Violin Memory Gateways. This software supports Fibre Channel and iSCSI targets. The steps to configure a vSHARE Violin Memory Gateway differ slightly depending on the transport protocol (Fibre Channel protocol or iSCSI) and the level of security required. However, the basic process for all transport protocols, includes the following tasks: 1. Configure a container on a Memory Array. For more information, see "Configuring Storage

Containers" on page 180. 2. Configure target ports used to control access to LUNs. For more information, see "Configuring

Target Ports" on page 182.

535-0008-00 Rev 10


177


Note: If you are using Fibre Channel, the target ports are automatically configured when you create the storage container on the Memory Array. If using iSCSI, the target ports must be configured manually. 3. Configure initiator groups and initiators. For more information, see Manage Initiators on page

69. 4. Configure LUNs. For more information, see Manage LUNs on page 59. 5. Export LUNs. For more information, see Manage LUNs on page 59.

The following flowchart provides a visual overview of the process for configuring the Memory Gateway for vSHARE block storage.

Figure 4.2 vSHARE Deployment Flowchart

The vSHARE configuration process consists of the following procedures:

Note: Configuring target ports and initiator groups are optional steps in the vSHARE configuration process. Organizations that do not want to use LUN-masking can control access to LUNs through ports or switches. LUN masking is an authorization process used to make a logical unit number available to some hosts and unavailable to others. 1. Configuring Storage Containers on page 180. This step initializes the Violin Memory Array to

manage block storage and create one or more storage containers (partitions) to manage the LUNs. To format the Violin Memory Array from vSHARE, the media init command is required. This command enables you to initialize one or more Memory Arrays to support block storage. Once a Memory Array has been initialized to support block storage, you may create one or more storage containers on each Memory Array, which may be used to store and manage LUNs. 2.

(Optional) Configuring Target Ports on page 182. This step configures the target ports so they can control access to LUNs. Configuring target ports is useful for security and bandwidth management.

178


535-0008-00 Rev 10

AIX Compatibility NACA Bit Support

3.

(Optional) Configuring Initiator Groups on page 187. This step configures the initiator groups, adding one or more initiators to each initiator group. You can restrict access to LUNs by initiator group, or on a per-initiator basis.

4. Working with LUNs on page 190.

LUNs are created within a storage container that comprises a particular Violin Memory Array. A LUN inherits the attributes of the container (Violin Memory Array) in which it is created.

Note: As of this release, one container comprises a Violin Memory Array. 5. Exporting LUNs on page 193. This required step exports the LUNs to the initiator groups or

initiators via target ports. Only those initiator groups or initiators to which the LUN is exported may access the LUN. Access may be further restricted to a specific target port.

AIX Compatibility NACA Bit Support LUNs with NACA (Normal Auto Contingent Allegiance) option enabled support a NACA bit set to one, the CDB control byte of the host SCSI command. This in turn enables AIX to use a queue depth of more than one. This provides a performance enhancement for IBM AIX systems only.

Caution: This feature is not recommended for use with operating systems other than AIX.

You can turn on (1) this NACA bit support when you create a new LUN, as demonstrated in the following task. Optionally, you can turn on the NACA option after a LUN is created. This feature is turned off (0) by default. For instructions on how to turn on NACA through the Violin Web interface, see described in Creating and Deleting LUNs on page 62. You have the option of modifying this feature through the Web interface, as described in Editing LUNs on page 64. To enable NACA bit support when creating a new LUN, do the following: 1. Connect to the Violin Memory Gateway through a terminal port, such as PuTTY. 2. Enter Configure mode, using the following commands.

> enable # configure terminal 3. Create a new LUN and enable NACA.

# lun create container name size 5Gb naca Caution: After the NACA option is enabled, you should only export the LUN to an AIX host.

535-0008-00 Rev 10


179


Configuring Storage Containers The first step in configuring a vSHARE Memory Gateway is to format the Memory Arrays and create the containers needed to manage LUNs in the target storage on each memory array. In vSHARE, LUNs are addressable subsets of the flash memory within a Violin Memory Array. Every LUN created and managed by vSHARE is created within the storage container that comprises the Array.

Note: As of this release, one container comprises a Violin Memory Array.

When a LUN is created, it is assigned to a specific container and inherits the attributes of that container. Configuring a Storage Container uses the following process: 1. Initializing Violin Memory Arrays on page 180 2. Viewing Containers on page 182

Initializing Violin Memory Arrays To configure a Violin Memory Array to support block storage, you must first initialize the Violin Memory Array using the media init command. This section introduces the media init command and how to use it in the following topics:

• •

Command Syntax and Parameters on page 180 Configuring Block Storage on page 181

Command Syntax and Parameters This section covers the command syntax for the media init command and lists and explains the available parameters. Syntax

media init device type [block] [force][name] WARNING! Using the force option of the media init command (media init force) destroys the existing data on an array. You should not use this command on a previously initialized array without first backing up the data on the array.

180


535-0008-00 Rev 10

Configuring Storage Containers

Parameters

device

type name

force

The parameter indicates the array to be initialized and is a string of the form: ataVIOLIN_MEMORY_ARRAY_xxxxxxxxxxxxxx The type parameter defines media. The name parameter is used to name the partition. The default is to use the last 14 characters of the Violin Memory Array serial number. To use a different name, add this parameter followed by a custom name of choice. The name can be up to 14 alphanumeric characters in length. The force parameter is used to force initialization of a Violin Memory Array that has already been initialized.

Configuring Block Storage This section demonstrates the process of initializing a Violin Memory Array for block storage. To initialize the Violin Memory Array for block storage, complete the following steps: 1. Connect to the Memory Gateway through a terminal port, such as PuTTY. 2. Enter Configure mode, using the following commands.

> enable # configure terminal (config) # 3. Enter the media init command to initialize the Violin Memory Array.

(config) # media init device ata-VIOLIN_MEMORY_ARRAY_23108R00000043 type block This example generated the container ID: 23108R00000043. 4. (Optional) Use the name parameter to create a more intuitive name for the container ID. In the

following example, the generated container ID is changed to FASTBOX1: (config) # media init device ata-VIOLIN_MEMORY_ARRAY_23108R00000043 type block name FASTBOX1 5. Confirm that you want to initialize the Violin Memory Array.

Really initialize ata-VIOLIN_MEMORY_ARRAY_FASTBOX1 (all data will be lost)? [no] y A message confirms that the Violin Memory Array was successfully initialized. media init completed successfully SM076 [MY-vSHARE00: standby] (config) #

535-0008-00 Rev 10


181


Viewing Containers Once you have initialized the Violin Memory Array, you can see the container that was created. This section covers how to use the show containers command to view existing containers. Before a block media device can be used for block storage, it must be enabled. In the following task, you use the media block id enable command to enable the device. To view existing containers, complete the following steps: 1. Log on as an administrator to the Memory Gateway attached to Port 1 of the Violin Memory

Array. 2. Use the configure terminal command to enter Configure mode.

# configure terminal (config) # 3. Enter the media block id all enable command to enable all block containers.

(config) # media (config) # media block id all enable The block parameter ensures that you enable vSHARE. The all parameter ensures that every block container is enabled. 4. To verify the containers were created, enter the show containers command. Information

about the containers displays. > enable HR-mga [cluster: master] # show containers Container LUNs Total Free Prov Ports HostA HostB Status Encrypted --------------------------------------------------------------------HR2013 30 10356G 9745G 459G A,B HR-mga HR-mgb HA No

Configuring Target Ports The second step in configuring a vSHARE Memory Gateway is to configure the target ports. The target ports control access to a particular LUN, which may be useful for both security and bandwidth management. This section covers the following topics:

• • •

182

Supported Target Ports on page 183 Configuring Fibre Channel and InfiniBand Target Ports on page 183 Configuring iSCSI Target Ports on page 184


535-0008-00 Rev 10

Configuring Target Ports

Supported Target Ports In a vSHARE system, the Violin Memory Arrays are the targets and the host system (for example, one or more database servers or application servers) are the initiators. The storage systems have storage target devices, LUNs, which the hosts access through the Memory Gateway. vSHARE supports the following target port types:

• • •

Fibre Channel InfiniBand SRP iSCSI

Every target is a specific port on a hardware Fibre Channel host bus adapter (HBA), InfiniBand host channel adapter (HCA), or a software-configured target for iSCSI.

•

For Fibre Channel or InfiniBand, the target ports are present when the Memory Gateway boots, if the HCAs or HBAs are installed on the system. Creating a container is not necessary.

•

For iSCSI, the target ports must be configured manually.

Note: The SCSI RDMA Protocol (SRP) is an industry standard protocol for utilizing block storage

devices over an InfiniBand fabric.

A LUN may be exported through multiple target ports of different types. Active-active multipathing is supported.

Configuring Fibre Channel and InfiniBand Target Ports For Fibre Channel or InfiniBand, the target ports are present when the Memory Gateway boots, if the HCAs or HBAs are installed on the system. To confirm that the vSHARE system can see the target ports prior to creating the initiator groups, use the show targets command. The show targets command enables you to view all target ports, whether Fibre Channel, InfiniBand SRP, or iSCSI.

Note: Mellanox ConnectX-2 InfiniBand HCA is supported in release G5.2.0. Be advised that the two ports on Mellanox HCAs are not entirely separate entities with respect to target port masking. Port steering for hca-a1 also uses hca-a2.

show targets Command The show targets command lists all of the target ports and their addresses (wwn/iqn), filtered by various options. Syntax

show targets [node ] [hostname ] [protocol ] [id ] [sessions] [detail]

535-0008-00 Rev 10


183


Parameters

[node ]

Displays targets on node ID.

[hostname ]

Displays targets on hostname.

[protocol ]

Displays targets using protocols FC, IB, or iSCSI.

[id ]

Displays a specific target.

[sessions]

Displays targets including sessions.

[detail]

Displays in detail.

Configuring iSCSI Target Ports This section describes how to configure iSCSI target ports, as well as providing information about the commands that are used. iSCSI target commands enable you to define iSCSI-specific options. By default, a single target corresponding to the host name of the Memory Gateway is available after the initial setup. Users wishing to create multiple targets per hosting node may use a separate command to create targets for each interface. SANs using iSCSI can operate over Ethernet. The Ethernet network can either be a dedicated or shared network. If the SAN is using a dedicated Ethernet network, the security of an iSCSI network is as good as that of a Fibre Channel network. If the SAN is running over a shared network, you must take steps to ensure the same level of security.

Note: In the case of an HA pair, when you create a target and bind it to an interface on one node,

the target must also be a bound to the (same) interface on the other node of the HA pair.

iSCSI Target Port Configuration To configure an iSCSI target port, complete the following steps: 1. From the CLI, enter into Enable mode using the enable command. 2. Every target port must be bound to an interface. Run the show interfaces command to

view the available interfaces. # show interfaces brief Using the brief parameter ensures that the CLI displays only high level information about the available interfaces. You will need the name of the interface (for example, eth3), its and to configure the interface in Step 3 of this procedure. 3. Enter into Configure mode using the configure terminal command and then run the

interface ip address command to configure a specific interface. # configure terminal (config) # interface ip address

184


535-0008-00 Rev 10

Configuring Target Ports

Typically, you should configure one of the four 10GbE interfaces identified as eth3, eth4, eth5, or eth6 to act as the iSCSI target port. In Step 6 of this procedure, you will bind the iSCSI target to the IP address of the configured interface. 4. Run the iscsi enable command to enable the iSCSI target ports.

(config) # iscsi enable The iscsi enable command creates a single iSCSI target called “iSCSI.” The iscsi enable command may, in conjunction with an optional [no] parameter, be used to disable iSCSI target ports. For more information see iscsi enable Command on page 185, 5. Run the iscsi target create command to create and name a target.

(config) # iscsi target create FINANCE The command creates an iSCSI target named “FINANCE.” Access to this port and the LUNs visible through this port may be restricted. Additional targets and target names enable you to define physical or logical separation of groups of users of the system. 6. Run the iscsi target bind command to bind the target port to one or more configured

interfaces. Interfaces are identified by their IP addresses. (config) # iscsi target bind to You may bind the iSCSI target port to multiple IP addresses. 7. Run the show targets command to confirm that the target ports have been correctly

configured. (config) # show targets Node Hostname Protocol Target Enab Address ---- --------------- -------- ---------- ---- -------4 iscsi yes iqn.200402.com.vmem:: The address of the iSCSI target consists of three elements: the IQN (iqn.200402.com.vmem), the , and the . The IQN consists of the literal (iqn), the date (yyyy-mm), and the reversed domain name of the vendor, Violin Memory. A colon prefixes the node_name. The default node_name is generated from the hostname. A second colon prefixes the user-defined target_name. iscsi enable Command The iscsi enable command may be used to enable or disable iSCSI target ports. Syntax

[no] iscsi enable

535-0008-00 Rev 10


185


iscsi target create Command The iscsi target create command enables you to create and name a target. Additional target names may be created if needed. The target name consists of up to 32 alphanumeric characters. Syntax

[no] iscsi target create Syntax Description

A user-defined name for the target. “iSCSI” is the default name used for the target automatically created when you use the iscsi enable command.

show targets Command The show targets command lists all of the target ports and their addresses (wwn/iqn), filtered by various options. show targets [node ] [hostname ] [protocol ] [id ] [sessions] [detail] Syntax Description

[node ]

Displays the targets on a specified node (Memory Gateway) in the cluster.

[hostname ]

Displays targets on hostname.

[protocol ]

Displays targets using protocols FC, IB, or iSCSI.

[id ]

Displays a specific target.

[sessions]

Displays the number of SCSI sessions are active on a target. The session parameter shows the initiator and the assigned LUN ID for each session.

[detail]

Displays detailed information about a target.

iscsi target bind Command The iscsi target bind command enables you to bind an iSCSI target to one or more IP addresses. The command limits visibility of the target to initiators which have access to those IP addresses.

Note: For an HA pair, when you create a target and bind it to an interface on one node, the target

must also be a bound to the (same) interface on the other node of the HA pair.

Syntax

[no] iscsi target bind name to [to ]*

186


535-0008-00 Rev 10

Configuring Initiator Groups

Syntax Description

A user-defined name for the target.

The IP address on a node on which iSCSI targets are enabled.

[to ]

The target may be bound to multiple IP addresses.

Configuring Initiator Groups The third step in configuring the vSHARE system is to configure the initiator groups (igroups) which identify the initiators (hosts) that are allowed to access a LUN or set of LUNs. Every initiator has a protocol-specific identifier depending on whether the iSCSI, Fibre Channel or InfiniBand protocol is used to access storage.

•

iSCSI initiator names have several formats: “iqn...”, “naa...”, “eui...” of which “iqn” names are most prevalent. For example, iqn.2004-02.com.vmem.

•

Fibre Channel initiators are identified by World-Wide Names (WWN). Fibre Channel initiators are fixed by the appropriate HBA port.

•

InfiniBand initiator names have a guid identifier format such as, guid.1234:5678:9abc:def0.

You can include an initiator group within another initiator group to create a nested initiator group (nested igroup). A nested igroup can contain one or more igroups. For more information on nested initiator groups, see Working with Nested Initiator Groups on page 194 and Managing Nested Initiator Groups on page 213. Initiator Group Configuration If you intend on using nested initiator groups, it is recommended that you plan the nested igroups hierarchy following the guidelines described in Working with Nested Initiator Groups on page 194 and Managing Nested Initiator Groups on page 213. To configure an initiator group, complete the following steps: 1. From the CLI, use the enable command (to enter Enable mode), and then the configure

terminal command (to enter Configure mode). > enable # configure terminal (config) # 2. Use the igroup name command to create an initiator group, as shown in the following

example. (config) # igroup create name 3. Add one or more initiators to the igroup, using the igroup addto command. The

initiator_name must be a valid iqn (iSCSI) or wwn (Fibre Channel), or guid (InfiniBand) name, such as, iqn.2004-02.com.vmem:initiator1.

535-0008-00 Rev 10


187


In the following example, an initiator named iqn.2004-02.com.vmem:initiator1 is added to the igroup. A colon separates the IQN from the user-defined name for this initiator. (config) # igroup addto name initiators iqn.200402.com.vmem:initiator1 An igroup can contain different types of initiators, Fibre Channel, InfiniBand, and iSCSI. In the following example, two Fibre Channel initiators are added to the igroup that contains an iSCSI initiator. (config) # igroup addto name initiators wwn.21:00:00:1b:32:9a:18:65 wwn.21:01:00:1b:32:ba:18:65 Fibre Channel initiator identifiers (WWN) are fixed by the HBA port, and InfiniBand initiator identifiers (guid) are fixed by the HCA port. If you want to restrict LUN access to a particular Fibre Channel or InfiniBand initiator, you must get the appropriate the WWN from the Fibre Channel HBAs, or the GUID from the InfiniBand HCA, on the client machine. 4. To verify that the initiators were added to the initiator group, return to the CLI and enter the

show igroups command. (config) # exit # show igroups Initiator group: #Initiators: 4 iqn.2004-02.com.vmem:initiator1 iqn.2004-02.com.vmem:initiator2 wwn.21:00:00:1b:32:9a:18:65 wwn.21:01:00:1b:32:ba:18:65 The show igroups command shows all initiator groups, the number of initiators within each group, and the IQN or WWN of those initiators. igroup create Command The igroup create command creates an initiator group and, optionally, one or more iSCSI or Fibre Channel initiators. iSCSI initiator identifiers (IQNs) are user-defined. Fibre Channel initiator identifiers (WWNs) are generated automatically by an HBA. Syntax

[no] igroup create name Syntax Description

name

188

The name is alphanumeric only.


535-0008-00 Rev 10

Configuring Initiator Groups

Examples

The following examples demonstrate how the igroup create command can be used. to manage initiator groups and initiators. The following command creates an initiator group named FINANCE: igroup create name FINANCE The Cisco-style no prefix may be used to delete initiator groups or initiators. In the following example, the command deletes the igroup FINANCE. no igroup create name FINANCE igroup addto Command The igroup addto command adds one or more initiators to a group. Syntax

[no] igroup addto initiators [initiator_name …]

igroup_name

The name is alphanumeric only.

initiator_name

The initiator_name must be prefixed with “wwn.” (FC initiators), “guid.” (InfiniBand initiators), or “iqn.” or “naa.” or “eui.” (iSCSI).

The following command adds three initiators to the initiator group named FINANCE: igroup addto FINANCE initiators wwn.20:20:20:20:20:20:20:20 wwn.20:20:20:20:20:20:20:21 wwn.20:20:20:20:20:20:20:22 The following command removes the initiator wwn.20:20:20:20:20:20:20:20 from igroup FINANCE and preserves the other two initiators (wwn.20:20:20:20:20:20:20:21 wwn.20:20:20:20:20:20:20:22). no igroup addto FINANCE initiators wwn.20:20:20:20:20:20:20:20

535-0008-00 Rev 10


189


Working with LUNs In vSHARE, LUNs are addressable subsets of the flash memory within the container that comprises a Violin Memory Array. In many systems, a LUN may be referred to as a volume or logical unit. This section covers the following topics:

• • • • •

Creating LUNs on page 190 Viewing LUNs and Containers on page 192 Resizing LUNs on page 192 Resizing LUNs on page 192 Exporting LUNs on page 193

Note: As of this release, one container comprises a Violin Memory Array.

Creating LUNs On the host system (initiator), the LUN appears as a local SCSI disk. The host may format and partition the LUN. The target, or storage system, sees the contents of the LUN as a set of blocks of arbitrary data. Every LUN is created within a container, and each container comprises a Violin Memory Array. vSHARE implements LUNs at a minimum of 1GB increments with the ability to configure up to 1024 LUNs per Memory Gateway. LUN Create Command The lun create command enables you to create a LUN within a specified storage container. Command Syntax

[no] lun create container name size [ | equal] [quantity ] [nozero] [readonly] [startnum ] [blksize 512 | 4096] [offline] alua naca Arguments and Options

The table shows syntax for LUN commands. name

Defines the name of LUN to be created; used as a prefix if the quantity of LUNs created is greater than one.

size

Defines the size of each LUN in GB, or “equal” to divide the free space of the container equally among the new LUNs. A value of 0 (zero) may be used in place of “equal.”

container quantity

190

Identifies the name of the container in which the LUN is created. Defines the number of LUNs to be created; the default is 1. If the number is greater than 1, the names of the newly created LUNs will be appended with an index number from startnum up to the number of LUNs created.


535-0008-00 Rev 10

Working with LUNs

nozero

Specify to avoid zeroing-out the data on the LUNs, which could take a long time for large LUNs.

readonly

Creates the LUNs in read-only mode.

startnum

Defines starting index number for newly created LUNs. The default is 1. Defines the logical block size for the created LUNs; the default is 512, can be set to 512 or 4096. Not all systems can handle 4096byte sizes. Use 512 if you are in doubt.

blksize

offline

Creates the LUNs in offline mode.

alua

Turns on ALUA for a LUN. For more information on ALUA, see Optimizing Performance with ALUA on page 257.

naca

Turns on NACA for a LUN. For more information on NACA, see AIX Compatibility NACA Bit Support on page 179.

LUN Create Example The following example demonstrates creating 20 LUNs of equal size in the container. Use the nozero option to ensure that the data is not zeroed-out, which could take considerable time depending on the size of the container.

Note: You can create up to 1024 LUNs per container.

To create a LUN, complete the following steps: 1. From the CLI, enter into Enable mode using the enable command and then Configure mode

using the configure terminal command. > enable # configure terminal (config) # 2. All LUNs are created and managed within a container. To view the available containers, use

the show containers command. > enable # configure terminal # show containers Check the size of the containers. LUNs are implemented in 1GB increments; the size of the container may restrict the number of LUNs that may be created. 3. Enter the lun create command to create LUNs within a container.

# lun create container name size equal quantity 20 nozero

535-0008-00 Rev 10


191


Setting LUN Access Parameters The lun set command allows you to set the parameters of a LUN in a container. For example, you can set the LUN as read-only, as well as enable and disable the export of a LUN through one or both Memory Gateways. The lun set command is the only command usable after a LUN has been created except the no lun create … readonly command. Command Syntax

[no] lun set container name readonly

Viewing LUNs and Containers This section covers the commands for viewing data for LUNs and the container in which they reside. Viewing LUNs The show luns command allows you to view various types of information about a LUN or LUNs. Command Syntax

show luns [container] [name] [count] [reserved] [serial] [sessions] Arguments and Options

container

Displays only the LUNs within a specific container.

name

Displays a specific LUN.

count

Displays LUNs with active sessions and their count.

reserved

Displays LUNs with SCSI reservations.

serial

Displays the serial numbers for each LUN.

sessions

Displays the sessions accessing a particular LUN.

Viewing Containers The show containers command displays all available containers. Command Syntax

show containers

Resizing LUNs You can increase the size of an existing LUN by specifying a the larger size in Gigabytes (GB), or by specifying an amount (GB) to add to the current size. Decreasing the size of an existing LUN is not supported. For a smaller size LUN, create a new LUN of the desired size.

192


535-0008-00 Rev 10

Working with LUNs

WARNING! Existing client side file systems that might be stored on the LUN are not extended as part of the resize operation. Increasing the size of an existing client side file system must be done on the client. Some client systems do not detect a LUN size change until they reboot.

Specifying a Larger Size for a LUN The following command increases the size of an existing LUN by specifying the new size in Gigabytes. lun resize container name size Specifying Additional Gigabytes for a LUN The following command increases the size of an existing LUN by specifying the amount (GBs) to increase the current size. lun resize container name size +

Exporting LUNs The fifth step in configuring the vSHARE Memory Gateway is to export the LUNs to an initiator. The LUN must be exported before it may be accessed over block protocols.

Note: vSHARE LUNs are defined and saved within the Arrays themselves. vSHARE LUN export

records are, however, managed and stored within the vCLUSTER configuration database owned by the master node.

Access to LUNs may be restricted to a specific initiator group, initiator, or target port when the LUN is exported using the lun export command.

•

The optional [igroup | initiator] parameter identifies the initiator group or initiator that may access the LUN.

•

The optional [port] parameter identifies the target that may access the LUN.

A LUN ID is a SCSI identifier which may be used to differentiate between devices on the same target port. By default, vSHARE automatically assigns LUN IDs to LUNs when they are exported to an initiator group or initiator.



vSHARE optionally enables you to assign a special, user-defined LUN ID to a vSHARE LUN when you export LUNs to an initiator group or initiator. User-defined LUN IDs may make it easier differentiate between LUNs. If you do assign user-defined LUN ID, Violin Memory recommends that you specify a number below 255 as some operating systems (for example, Windows) will only discover LUN IDs between 0 and 254. LUNs cannot be exported if they are assigned an existing LUN ID; an error message is returned and the export fails.

535-0008-00 Rev 10


193


Working with Nested Initiator Groups A nested initiator group (nested igroup) is an igroup that contains one or more igroups. This concept is similar to a folder that contains other folders, or subfolders. You can create up to three hierarchy levels in a nested igroup. For information on creating nested initiator groups, see Creating, Modifying, and Deleting Nested Initiator Groups on page 72 and Managing Nested Initiator Groups on page 213.

Guidelines for Exporting LUNs The following guidelines apply when exporting LUNs.

• •

Each LUN ID (LUN name) must be unique to avoid the chance of conflicts.

•

Changing the LUN ID (LUN name) after a LUN has been exported is not recommended, as traffic is disrupted during the change. The affect on the client depends on how it handles LUN ID changes.

•

Re-exporting a LUN with an auto-assigned LUN ID may cause the system to assign a new LUN ID to the export, which may disrupt the existing I/O traffic. The same holds true for manual LUN ID changes.

•

You can export LUNs to different igroups on different targets, even when they have initiators in common.

Exporting a LUN to different targets using a different LUN ID (LUN name) for each target can result in conflicts.

lun export Command The lun export command enables you to export the LUNs within a specific container to an initiator or initiator group. A single LUN may be exported multiple times through different target ports (multipathing) or to different initiators (clustered access). Syntax

[no] lun export container name [lunid][to ]... [using ]

194


535-0008-00 Rev 10

OpenVMS Support

Arguments and Options

lun_name

The lun_name could be wildcard specified, such as FINANCE*

container

The name of the container

name

The name of the LUN

lunid

Designates the LUN ID assigned to an exported LUN. By default, LUN IDs are automatically assigned by vSHARE. User-defined LUN IDs may be specified to track specific LUNs.

igroup

The name of the igroup. If no igroup or initiator is specified, all initiators will have access to the LUN.

initiator

The IQN (iSCSI) or WWN (Fibre Channel) name of the initiator.

port

The port is iSCSI portal (i.e. 10.10.0.14) or Fibre Channel wwn ID.

OpenVMS Support OpenVMS (Open Virtual Memory System) is a multi-user, multiprocessing virtual memory-based operating system (OS). OpenVMS servers require that LUNs have a valid Unique Device Identifier (UDID) for discovery and usage by the operating environment. Support for OpenVMS provides the following capabilities:

• • •

Specifying the OpenVMS UDID at the time a LUN is created, or after a LUN is created Removing a LUN UDID from being reported Listing LUN UDID information

OpenVMS requires that a target array provide a LUN 0, which is seen as Disk Device Type (0x0). By default vSHARE provides LUN 0 as the SCSI Enclosure Services Device Type (0x0D). OpenVMS specifies LUN 0 as a controller device, with other LUNs appearing on other LUN identifiers.

Creating a LUN and Specifying the UDID This section demonstrates how to specify the LUN ID to be used during export, thereby overriding the auto-assignment that begins at LUN ID 1. In the following task, you create a 1 Gb LUN, set an OpenVMS UDID, and then export the LUN to all initiators as LUN 0. The disk LUN 0 is for discovery only and should be as small as possible, hence the 1 GB LUN.

535-0008-00 Rev 10


195


To create a LUN and then export it as LUN 0, do the following: 1. Log in as Admin and connect to the Violin Memory Gateway through a terminal port, such as

PuTTY. 2. Enable configure mode, using the following commands.

> enable # configure terminal (config) # 3. Create a 1 Gb LUN in a container.

# lun create container name size 1 Note: The disk LUN 0 is for discovery only and should be as small as possible, (1 GB is recommended). 4. Set the unique device identifier (UDID) for the LUN, and then confirm change.

# lun set container name udid 55 Data discription will occur. Really change UDID for LUN lun_name [no] y Success 5. Export the newly created LUN as LUN 0.

# lun export container name lunid 0 6. Save the configuration. # write mem 7. Verify that the LUN was created.

# show luns udid Container: VIOLIN LUN UDID ___________________________ MyLUN

196

55


535-0008-00 Rev 10

Working with LUN Groups

Setting and Removing a LUN UDID This section demonstrates how to specify and remove unique a UDID for a LUNs in a container. To set or remove a LUN UDID, do the following: 1. Log in as Admin and connect to the Violin Memory Gateway through a terminal port, such as

PuTTY. 2. Enable configure mode, using the following commands.

> enable # configure terminal (config) # 3. To set the UDID of a LUN in a container (array), use the following command.

# lun set container name udid 4. To remove the UDID for a LUNs in a container, use the following command.

# no lun set container name udid

Working with LUN Groups All LUNs in a LUN group must belong to the same array. A LUN group inherently has the same functionality as a single LUN. Actions to a LUN group are applied to all LUNs that are members of the LUN group. All the LUNs used by a specific application can be grouped together into a LUN group. You can have a maximum of eight LUNs in a LUN group.

Note: A LUN group is a management object only, and does not impact the data path.


• • • • •

Creating LUN Groups on page 197 Deleting LUN Groups on page 198 Modifying LUN Groups on page 198 Exporting a LUN Group on page 199 Displaying LUN Groups on page 200

Creating LUN Groups You can create a LUN group with one or more LUNs. When you perform an action on a LUN group, that action is applied to all the LUNs that are members of the group. A LUN group typically consists of LUNs that are used by a single application. For example, a database application may require two or more LUNs for data and logs.

535-0008-00 Rev 10


197


Note: You can create a LUN group with a maximum of eight LUNs.

The following guidelines apply:

• • • •

A LUN may be part of multiple LUN groups. A LUN group can contain a minimum of one LUN and a maximum of eight LUNs. A LUN group name must contain alphanumeric characters. A LUN group name must be unique within a container.

Substitute LUN names for the variables in the command above. You must have one LUN to create a LUN group, and optionally can specify seven other LUNs as members. # lungroup create container name lun lun [ lun lun ... lun ]

Deleting LUN Groups You can delete a LUN group without affecting the LUNs belonging to the group. Deleting a LUN group in effect only deletes the relationship that binds the member LUNs together as an object.

Note: You must delete the snap groups (snapshots of a LUN group) associated with a LUN group before you are allowed to delete the LUN group. Snapshots that belong to a group cannot be deleted individually. You must delete all the snapshots in a snap group.

# no lungroup create container name

Modifying LUN Groups You can change a LUN group name, as well as the LUNs that are members of the group. You can add new LUNs to a LUN group, and remove existing LUN members. This function only impacts the management of the LUN group (object), and does not create or delete actual LUNs.

Note: If a LUN is deleted, it is automatically removed from any LUN groups of which it is a member. If deleting a LUN results in a LUN group with zero members, then that LUN group is automatically deleted.

Adding LUNs to a LUN Group Substitute appropriate names for the variables in the following command. # lungroup addto container name lun Note: A LUN group can have a maximum of eight LUNs.

198


535-0008-00 Rev 10

Working with LUN Groups

Deleting LUNs from a LUN Group Substitute LUN names for the variables in the following command. # no lungroup addto container name lun lun [ lun lun ... lun ] Note: You can delete up to seven LUNs from a LUN group. One LUN must remain to maintain

the group. If you delete all LUNs from a LUN group, the LUN group is automatically deleted.

Renaming a LUN Group To rename a LUN group, specify the container in which the LUN resides, the existing name (old name) of the LUN followed by the new name of the LUN.

Note: Renaming a LUN group with exported LUNs is not recommended, as traffic can be

disrupted. For more information, see Guidelines for Exporting LUNs on page 194. # lungroup rename container name to

Exporting a LUN Group You can export a LUN group and all the LUNs contained within the LUN group, using the following options. — to: exports a LUN group to an initiator or initiator group — using: exports a LUN group using a target port — description: allows you to add a description to identify the LUN group Rules for exporting LUNs also apply to LUN groups. For more information, see Guidelines for Exporting LUNs on page 194.

Exporting a LUN Group to an Initiator Port The following command exports a LUN group to a specified initiator port. # lungroup export container name to

Exporting a LUN Group Using Specific Ports The following command exports a LUN group using specified ports. # lungroup export container name using

535-0008-00 Rev 10


199


Adding a Description to a LUN Group # lungroup set container name description This command is adds a unique description to the LUN group.

Displaying LUN Groups You can display a list of all LUN groups on a system. Each LUN group name and the LUNs present in each group are shown. When a LUN name changes, the LUN group list shows the new name. You can specify a container, and only show the LUN groups in that container. The following command displays a list of LUN groups and their individual members. # show lungroups [container ]

# show lungroups Container: MyContainer Container: 41238F00608 LUN Group LUNs ---------------------------------------------------------------------lung1 lun1,lun2,lun3

# show lungroups container 41238F00608 Container: 41238F00608 LUN Group LUNs ---------------------------------------------------------------------lung1 lun1,lun2,lun3

# show lungroups container 41238F00608 name lung1 Container: 41238F00608 LUN Group LUNs ---------------------------------------------------------------------lung1 lun1,lun2,lun3 Total Allocated Size: 0G Lun Allocated/Provisioned Size: 0G/3G Snapgroup Protected/Unprotected: 0G/0G

200


535-0008-00 Rev 10

CHAPTER 5

vSHARE Management

This chapter describes the management of a Violin Memory Gateway configured for vSHARE block storage using the command line interface (CLI) and Violin Web Interface in four sections:

• • • • • •

Verifying Connections and Performance on page 201 Block Storage Media Management on page 203 vSHARE Block Storage Management Commands on page 211 Managing Block Storage in the Violin Web Interface on page 217 Email Alert Consolidation on page 217 Unified Logging Upload on page 219

Verifying Connections and Performance This section describes procedures and techniques which enable administrators to verify that LUNs are available and working over the required ports.

• •

Verifying Target Port Activity and Status on page 201 Verifying Container Activity and Status on page 203

Verifying Target Port Activity and Status The show targets and show interfaces commands enable you to verify the activity and status of target ports. Viewing Port Activity show targets [node ] [hostname ] [protocol ] [id ] [sessions] [detail] The show targets command lists all of the target ports and their addresses (wwn/iqn), filtered by various options. 535-0008-00 Rev 10


201

Chapter 5 - vSHARE Management

The session and detail options are particularly useful for understanding target port activity and status. The sessions option displays the number of SCSI sessions active on a target and the initiator and the assigned LUN ID for each session. SM075 [MY-vSHARE00: master] (config) # show targets protocol fc node 3 hostname SM076 sessions Node Hostname Protocol Target Enab Address ---- --------------- -------- ---------- ---- -----------------------3 SM076 fc hba-a1 yes wwn.21:00:00:1b:32:8f:24:5d Connected Sessions

3

SM076

fc

hba-a2

Connected Sessions

: wwn.21:00:00:1b:32:82:1e:50 : wwn.21:00:00:1b:32:94:13:32 : wwn.21:01:00:1b:32:a2:1e:50 : wwn.21:01:00:1b:32:af:24:5d : wwn.21:01:00:1b:32:b4:13:32 : wwn.21:fd:00:05:1e:09:ed:48 yes wwn.21:01:00:1b:32:af:24:5d : : : : :

wwn.21:00:00:1b:32:82:1e:50 wwn.21:00:00:1b:32:94:13:32 wwn.21:01:00:1b:32:a2:1e:50 wwn.21:01:00:1b:32:b4:13:32 wwn.21:fd:00:05:1e:09:ed:48

The detail option returns the network bindings for each target. SM075 [MY-vSHARE00: master] (config) # show targets protocol fc node 3 hostname SM076 detail Target: hba-a1 Node : 2 Hostname : qa-stein136-mga Protocol : fc Address : wwn.21:00:00:24:ff:38:54:b2 Enabled : yes State : Online Port Type Setting : Port Type : NPort (fabric via point-to-point) Speed Setting : Speed : 8 Gbit

For detailed information about this command, see Show Targets Command on page 216. Viewing Interface Status show interfaces [] [configured | brief] The show interfaces command may be used to view information about a specific interface or, if one is not named, all interfaces. By default, detailed information about the interface and its runtime

202


535-0008-00 Rev 10

Block Storage Media Management

state is given. The configured and brief options enable the user to specify which information is displayed in the CLI:

• •

If the configured option is selected, the configuration of the interface is displayed. If the brief option is selected, an abbreviated runtime state, with the interface statistics excluded, is displayed.

Verifying Container Activity and Status Viewing Media Statistics The show stats media command enables you to view read and write statistics for storage media. The show stats media command displays read and write rates (in MBs per second) for the current ten-second sample and averaged over the last five minutes. The show stats media command supports a continuous option which enables you to view a continually updated set of statistics. Viewing Containers The show containers command returns information about the storage containers on a Memory Gateway cluster including the node, container name, the number of LUNs, the space allocated, and the space free. HR-mga [cluster: master] # show containers Container LUNs Total Free Provisioned Ports HostnameA HostnameB Status Encrypted --------------------------------------------------------------------------------------Violin123 7 5332G 5245G 15G A,B violin123-mga violin123-mgb HA No

Block Storage Media Management This section describes commands for managing and monitoring block storage (vSHARE) media devices. This section contains the following topics:

• • • •

Show Media Commands on page 203 Media Block Enable Commands on page 207 Show Media Health Commands on page 209 Media Read and Write Statistics on page 210

Show Media Commands The show media commands enable you to view information about the storage media installed on a Memory Gateway node including the size, location, and status of all media devices. If no options are specified, the show media command returns information about vSHARE block storage media only and excludes all non-block storage media devices.

535-0008-00 Rev 10


203


The show media command may be defined by the following options, which may be used in various combinations: all

Display information for all media.

block

Display information about block storage media devices.

detail

Display detailed information.

freelist

Display media allocation freelist information for all devices including flash memory devices not allocated or formatted.

global

Display information about media on all Memory Gateways in the cluster.

health

Display status information and lifetime remaining for media devices. For more information on the health option, see Show Media Health Commands on page 209.

id

Display information for a specified media device or all media (id all).

The following show commands described in this section can be used in Configure or Enable mode. All other media management commands require Configure mode.

• • • • • •

show media show media all show media global show media block id all show media block id show media block detail

Note: For more information, see Command Modes on page 225.

show media

The show media command returns all media devices that can be used as block storage on the current Memory Gateway (node or module). This command shows a summary line for each device location, giving the size and status of that device. Location

Media ID

Model

Size

Status

--------------------------------------------------------------------unknown-00

Fender1002

Violin Memory Array

6895.75G

online

unknown-01

Strad130

Violin Memory Array

432.00G

online

The show media command is an alias for show media block id all command. Media used by the system for purposes other than block storage are not shown.

204


535-0008-00 Rev 10


show media all

The show media all command displays all media devices installed in a Memory Gateway. This command shows a summary line for each device location, giving the size and status (online/offline) of the installed media device. Location Media ID Model Size Status ---------------------------------------------------------------------Boot Disk

---

---

PCI 4

41148F00147

Violin Memory Array

74.53G

online

2671.00G

online

The Media ID and Model values are displayed only for media devices that are usable as block storage; otherwise, a string of three hyphens (---) is displayed instead. show media global

To view information about storage media devices on a cluster, connect to the master node of the cluster and add the global option to the show media commands. SM075 [MY-vSHARE00: master] # show media global Module 2: SM075 (10.10.0.37) Location

Media ID

Model

Size

Status

---------------------------------------------------------------------unknown-00

Fender1002

Violin Memory Array

6895.75G

online

unknown-01

Strad130

Violin Memory Array

432.00G

online

---------------------------------------------------------------------Module 3: SM076 (10.10.0.143) Location

Media ID

Model

Size

Status

---------------------------------------------------------------------PCI 2

Strad134

Violin Memory Array

399.75G

online

The show media all global command returns information about all media devices on the cluster including non-storage devices.

535-0008-00 Rev 10


205


show media block id all

The show media all block id all command displays the location, media ID, model, size, and status (online/offline) of all block storage media. SM075 [MY-vSHARE00: master] # show media block id all Location Media ID Model Size Status ---------------------------------------------------------------------unknown-00 Fender1002 Violin Memory Array 6895.75G online unknown-01 Strad130 Violin Memory Array 432.00G online

show media block id

Information about a particular block storage media device can be displayed by identifying the device with the block id option of the show media command: SM075 [MY-vSHARE00: master] # show media block id Strad130 Location Media ID Model Size Status ---------------------------------------------------------------------unknown-01 Strad130 Violin Memory Array 432.00G online The of specific media devices can be found in the output of the show media all or show media block id all commands. show media block detail

The detail option of the show media command returns detailed information about storage media. The detail option may be combined with any other show media command option in any order. SM076 [MY-vSHARE00: standby] (config) # show media block id all detail Location PCI 4:

206

Status:

online

Size:

2867964411904

Type:

block device

Firmware version:

6.0.0

Manufacturer model:

VIOLIN_MEMORY_ARRAY

Manufacturer serial no:

41148F00147

Device ID:

ata-VIOLIN_MEMORY_ARRAY_41148F00147

Violin model:

Violin Memory Array


535-0008-00 Rev 10


Violin serial no:

41148F00147

Violin Part number:

V6000

Violin Revision: Media Block Enable Commands Before a media device can be used for block storage, it must be enabled. This section describes commands for enabling and disabling block storage media devices on a vSHARE Memory Gateway. The enable commands described in this section can be used in Configure mode only. For more information, see Command Modes on page 225. Commands include:

• • • •

media block id enable no media block id enable media block id all enable no media block id all enable

media block id enable

The media block id enable command enables a specified block storage media device. The block option specifies a vSHARE block media device. In the following example, the media block id enable command is used to enable a single block storage media device: * gateway02 [cluster1: master] (config) # all Change block media VTMS0101-2 VTMS0101-4 * gateway02 [cluster1: master] (config) # enable * gateway02 [cluster1: master] (config) #

media block id ? state for all block devices

media block id VTMS0101-2 █

To enable all block storage media devices on a local Memory Gateway, use the all option as described in media block id all enable on page 208.

535-0008-00 Rev 10


207


no media block id enable

To disable a media device, use the no form of the media block id enable command. * gateway02 [cluster1: master] (config) # no media block id VTMS0101-2 enable Warning: this command will cause users of this media to reset and lose all content. Confirm command? [no] yes * gateway02 [cluster1: master] (config) # █

Caution: When a block media device is disabled, all applications using that device immediately stop using it, and will have space from other media allocated for them. This may be a disruptive operation.

To disable all block storage media devices on a local Memory Gateway, use the all option as described in no media block id all enable on page 208. media block id all enable

The media block id all enable command enables all block storage media devices on the local Memory Gateway. In the following example, the media block id all enable command enables all storage media devices on the local Memory Gateway: SM076 [MY-vSHARE00: standby] (config) # media block id all enable Enabling 23108R00000043 Enabling Strad134 no media block id all enable

To disable the use of all media devices in the Memory Gateway, use the no form of the above command in Configure mode.

Caution: When all media devices on a Memory Gateway are disabled, applications using those devices immediately stop using them and will have space allocated for them on other Memory Gateways. This may be a disruptive operation.

208


535-0008-00 Rev 10


In the following example, the no media block id all enable command disables all block storage media devices on the local Memory Gateway: * gateway02 [cluster1: master] (config) # no media block id all enable Warning: this command will cause users of this media to reset and lose all content. Confirm command? [no] yes Disabling VTMS0101-2 Disabling VTMS0101-4 * gateway02 [cluster1: master] (config) # █ Show Media Health Commands The health option of the show media command returns information about the status and expected lifetime of media devices. The option can be included with other options in various forms in any order.

• •

show media health show media health [all] [detail] [global]

The show media health commands display the location, media ID, model, life remaining, and status (or health) of media devices on the local Memory Gateway or on all of the Memory Gateways in the cluster. The commands can be used in Enable and Configure modes. Understanding Media Health Status

The status of the media device is defined by one of four values: ok

The device is running correctly and has plenty of expected life remaining.

unknown

The status cannot be determined.

warning

A S.M.A.R.T attribute of the device is within 5% of the usage threshold specified by the manufacturer.

critical

A S.M.A.R.T attribute is equal to or below its usage threshold. (This indicates a high likelihood of impending failure.)

Status is determined for each individual S.M.A.R.T attribute by comparing its current normalized value to a manufacturer-specified usage threshold. The detail option lists all of the S.M.A.R.T attributes and gives the status of each attribute. Without the detail option, show media health gives the worst-case status among all the device’s attributes. Normalized values begin at their maximum (usually 100 or 254) and decrease with use of the device, so they approach their specified thresholds from above. A warning status indicates that the value is close to its threshold, and a critical status indicates that it has reached or passed the threshold.

Note: Warning and critical status do not necessarily indicate any current problems; they only show that devices are near the manufacturer’s thresholds for expected lifetime performance. When a device reaches warning status, its detailed status should be monitored carefully.

535-0008-00 Rev 10


209


Caution: Critical status indicates that failures are likely in the near future. It is recommended that devices that reach critical status be disabled and replaced before they fail.

show media health

The show media health command returns information about the health of storage media devices on the local Memory Gateway. SM075 [MY-vSHARE00: master] # show media health Location Media Id Model Life remain Health ---------------------------------------------------------------------PCI 4 Violin123 Violin Memory Array 100.0% ok To view information about storage media on other nodes in the cluster, use the global option. The show media health command returns information about the block storage media only. To include non-storage media devices, use the show media all health command. show media health detail

The show media health detail command returns detailed information about block storage devices on the local Memory Gateway including all of the S.M.A.R.T attributes and provides the status of each attribute. In the following example, the block id option is used to identify a specific media device. SM075 [MY-vSHARE00: master] # show media health block id Strad130 detail Location unknown-01: Model: Violin Memory Array Est. life remaining: 96.0% Serial no: Strad130 Health status: ok SMART Attribute Type Raw value Norm Thres Status ------------------------------------------------------------------Reallocated Sector Ct Old age 0 100 0 ok Power On Hours Old age 0 100 0 ok Power Cycle Count Old age 0 100 0 ok Unsafe Shutdown Count Old age 0 100 0 ok Temperature Celsius Old age 54 54 0 ok Host Write Count Old age 0 200 0 ok Avail Reserved Space Old age 100 100 10 ok Media Read and Write Statistics The show stats media commands enable you to view read and write rates (in MB per second) for both the current ten-second sample and an average of the samples taken over the last five minutes.

210


535-0008-00 Rev 10

vSHARE Block Storage Management Commands

The show stats media command displays read and write statistics for block storage media devices in one or more nodes in the cluster. The scope of each command is the same as the corresponding show media command, described above. show stats media

The structure of the show stats media commands is identical to the show media commands described above, except that these commands also support a continuous option to display a continually updating view of the statistics. SM075 [MY-vSHARE00: master] # show stats media Media Stats Summary (sampled @10 secs) Read (bytes/s) Write (bytes/s) Read Latency (us) Write Latency (us) Media Current Last 5m Current Last 5m Current Last 5m Current Last 5m ----------------------------------------------------------------------------Violin123 0 21k 0 31k 112us 121us 74us 69us

If the continuous option is specified, the display of statistics is updated continually. Use CTRL+ C to exit the display. The command show stats media is an alias for show stats media block id all. The command returns information about storage media devices on the local Memory Gateway and omits all non-block storage media.

vSHARE Block Storage Management Commands This section describes CLI commands for configuring and managing block storage in five topics:

• • • • • •

Managing Containers on page 211 Supporting VMware VAAI on page 212 Managing iSCSI Targets on page 212 Managing Initiator Groups on page 213 Managing LUNs on page 214 Managing Targets on page 216

Managing Containers The show containers command returns information about the storage containers on a Memory Gateway cluster including the node, container name, the number of LUNs, the space allocated, and the space free. HR-mga [cluster: master] # show containers Container LUNs Total Free Prov Ports HostA HostB Status Encrypted --------------------------------------------------------------------HR2013 30 10356G 9745G 459G A,B HR-mga HR-mgb HA No

535-0008-00 Rev 10


211


Supporting VMware VAAI Violin vStorage API for Array Integration (VAAI) supports off-loading select VMware vSphere features from ESX to the array for processing. Full Copy and Block Zeroing are executed on the storage arrays to increase scalability and performance.

•

Full Copy—Enables the disk arrays to make deep copies of data without requiring ESX to access the data.

•

Block Zeroing—Allows ESX to send commands to zero blocks on the array, avoiding the need to send WRITE commands containing zeros to the array.

Note: By default VAAI is enabled on ESX. To disable VAAI, see the VMware knowledge base

article on Disabling the VAAI Functionality in ESX/ESXi.

Full Copy Sample Commands You can utilize Full Copy by cloning a VM or migrating a VM using Storage vMotion. You can perform these tasks on a vSphere Client using the graphical user interface (GUI). Block Zeroing Sample Commands You can utilize Block Zeroing by creating a zeroed-thick or eager zeroed-thick virtual disk on the vSphere Client using the graphical user interface.

Managing iSCSI Targets This section covers the following topics:

• • • •

Show iSCSI Targets on page 212 iSCSI Enable Command on page 213 iSCSI Target Create Command on page 213 iSCSI Target Bind Command on page 213

Show iSCSI Targets show iscsi The show iscsi command shows if iSCSI has been enabled, the number of bindings, the IP address, and interface for each binding. (config) # show iscsi iSCSI is enabled iSCSI Target: iscsi #Bindings: 4 10.1.9.82 (unknown) 192.168.1.101 (eth3) 192.168.2.100 (unknown) 192.168.3.101 (eth4)

212


535-0008-00 Rev 10


iSCSI Enable Command [no] iscsi enable The iscsi enable command may be used to enable or disable iSCSI target ports. iSCSI Target Create Command [no] iscsi target create The iscsi target create command enables you to create and name a target. Additional target names may be created if needed. The target name consists of up to 32 alphanumeric characters. A user-defined name for the target. The term “iSCSI” is the default name used for the target automatically created when you use the iscsi enable command. iSCSI Target Bind Command [no] iscsi target bind name to [to ]* The iscsi target bind command enables you to bind an iSCSI target to one or more IP addresses. The command limits visibility of the target to initiators which have access to those IP addresses.

Managing Initiator Groups You can link initiators into a group (igroup) and export LUNs to all the initiators in the group simultaneously.



Each initiator is a Fibre Channel WWN, iSCSI IQN or InfiniBand GUID address.

•

iSCSI initiatory identifier formats: iqn (most common), naa, eui For example, iqn.1994-05.com.redhat:xyz

•

Fibre Channel initiator identifier format: wwn Fibre Channel initiators are fixed by the appropriate HBA port, for example wwn.01:02:03:04:05:06:07:08

•

InfiniBand initiator identifier format: guid For example, guid.1234:5678:9abc:def0

Managing Nested Initiator Groups You can include an initiator group within another initiator group to create a nested initiator group (nested igroup). A nested igroup can contain one or more igroups. This concept is similar to a folder that contains other folders, or subfolders.

535-0008-00 Rev 10


213


Note: You can create up to three hierarchy levels within a nested igroup.

Best Practices for Nested igroups:

• •

Use a unique LUN ID (LUN name) for each LUN to avoid the chance of conflicts.

•

Refrain from exporting a LUN to different targets using a different LUN ID (LUN name) for the LUN for each target, as this practice can result in conflicts.

Plan the nested igroups hierarchy first. Then export the LUNs, allowing the system to autoassign LUN IDs to ensure they are each unique.

igroup Create Command [no] igroup create name initiators [initiator_name ...] The igroup create command creates an initiator group and, optionally, one or more iSCSI or Fibre Channel initiators. iSCSI initiator identifiers (IQNs) are user-defined. Fibre Channel initiator identifiers (WWNs) are generated automatically by an HBA. igroup addto Commands [no] igroup addto initiators [initiator_name …] The igroup addto command adds one or more initiators to a group.

Managing LUNs This section covers the following topics:

• • • •

Show LUNs Command on page 214 LUN Create Commands on page 215 LUN Set Commands on page 215 LUN Statistics Command on page 216

Show LUNs Command show luns [container] [name ] [sessions] The show luns command enables you to view LUNs. LUNs may be filtered by container, name, and session option.

214


535-0008-00 Rev 10


The show luns session command returns information about the sessions connected to LUNs including the target port and the session ID. Use the name option to identify a specific LUN. # show luns name Fender1002 sessions Container: HA-1 LUN Size RW Bksz Status Ports ------------------------------------------------------------------Fender1002 1024M rw 512 Single A Connected Sessions

: Port: hba-a1, Session: 21:00:00:1b:32:1b:8d:4b, LUN ID: : Port: hba-a2, Session: 21:00:00:1b:32:82:1e:50, LUN ID: : Port: hba-a2, Session: 21:01:00:1b:32:a2:1e:50, LUN ID: : Port: hba-a2, Session: 21:01:00:1b:32:b4:13:32, LUN ID: : Port: hba-a2, Session: 21:00:00:1b:32:94:13:32, LUN ID: : Port: hba-a2, Session: 21:01:00:1b:32:bd:4d:91, LUN ID: : Port: hba-a2, Session: 21:fd:00:05:1e:09:ed:48, LUN ID: : Port: hba-b1, Session: 21:01:00:1b:32:3b:8d:4b, LUN ID: : Port: hba-b2, Session: 21:00:00:1b:32:82:1e:50, LUN ID: : Port: hba-b2, Session: 21:01:00:1b:32:a2:1e:50, LUN ID: : Port: hba-b2, Session: 21:01:00:1b:32:b4:13:32, LUN ID: : Port: hba-b2, Session: 21:00:00:1b:32:94:13:32, LUN ID: : Port: hba-b2, Session: 21:fd:00:05:1e:09:ed:48, LUN ID: : Port: Procurve, Session: iqn.1994-05.com.redhat:Bow7,

1 1 1 1 1 1 1 1 1 1 1 1 1

LUN ID: 2 : Port: Procurve, Session: iqn.1994-05.com.redhat:Bow7_1, LUN ID: 2 : Port: Procurve, Session: iqn.1994-05.com.redhat:Bow7_2, LUN ID: 2 : Port: Procurve, Session: iqn.1994-05.com.redhat:Bow7_3, LUN ID: 2 To view a list of containers, enter show luns container ?. SM075 [MY-vSHARE00: master] (config) # show luns container ? Violin123 The name option enables you to view only those LUNs which are prefixed by a particular name. LUN Create Commands [no] lun create container name size [ | equal] [quantity ] [nozero] [readonly] [startnum ] [blocksize 512 | 4096] [offline] The lun create command enables you to create a LUN within a specified storage container. LUN Set Commands [no] lun set container name readonly

535-0008-00 Rev 10


215


The lun set command enables you to set a LUN as read-only. The lun set command is the only command usable after a LUN has been created except the no lun create … readonly command. LUN Statistics Command LUN statistics are accessed via the CLI with the following command: show stats lun [container ] [name ] [on global|on host ] [continuous] [detail] Use the container option to restrict statistics display to a single container. If it is omitted, statistics are shown for all the containers accessible to the hosts. Use the name option to get statistics for a single LUN. If it is committed, statistics for all LUNs are specified.

Note: Select shell wildcard patterns can be used in the field to match a set of LUNs. For example, you can use the asterisk (*) to match all characters that follow. Such as xyz* to match all LUN names that begin with xyz. Or, you can specify a range of LUN names using the [a-e] wildcard. Such as xyz-lun[2-4][0-9] to match all LUNs from xyz-lun20 to xyz-lun49.

Use the on global option to get statistics for data flowing to LUNs through all nodes in a cluster. Use the on host option to get statistics for data flowing to LUNs on a specific host. Use the continuous option to refresh displayed statistics every ten seconds. Use the detail option to display statistics in a verbose format.

Managing Targets This section covers the show targets command. Show Targets Command show targets [node ] [hostname ] [protocol ] [id ] [sessions] [detail] The show targets command displays all Fibre Channel or iSCSI targets. Use the hostname, protocol, and id parameters to filter the targets returned. The command returns the node, hostname, target port, status (enabled or not), and address (WWN or IQN). The detail parameter returns the network bindings for each target.

216


535-0008-00 Rev 10

Managing Block Storage in the Violin Web Interface

Managing Block Storage in the Violin Web Interface The Violin Web Interface provides you with tools for managing LUNs, initiator groups, and targets. For more information, see the following topics:

Section

Manage

Tasks and Topics

Searching, Sorting, and Performing LUN Operations on page 60 Creating and Deleting LUNs on page 62 Exporting LUNs, Initiator Groups, and Initiators on page 67 Creating and Deleting Initiator Groups on page 69 Adding or Deleting Initiators from a Group on page 71 Managing Transport Protocols on page 75

Monitor

Monitoring Container, LUN, and Port Data on page 108 Monitor LUNs on page 110 Monitor Targets on page 111

Table 5.1 Managing Block Storage Tasks in the Web Interface

Email Alert Consolidation You can manage email alerts, reducing the number of emails received, with the email consolidation feature. When you enable email consolidation, the first email to be sent is held for a default of 60 seconds. If additional emails are received within that time frame, they are merged into a single email before being sent out. You can customize the time delay, as well as the maximum number of events that are combined. For more information, see Customizing Email Consolidation on page 218. For information on how to see the current email settings, see Viewing Current Email Settings on page 218. Important points to remember about email consolidation:

•

If a system begins shutdown or if a maximum number of emails are staged (default 5), they are sent out automatically.

•

Emails are consolidated on a per-host basis. Emails from different hosts (acm-a, acm-b, mg-a, or mg-b) cannot be combined.

•

Log in to the ACM master or MG master node to configure email consolidation. The default is disabled.

Enabling and Disabling Email Consolidation You can enable and disable email consolidation using the following command. The default is disabled. Log in to the desired host (acm-a, acm-b, mg-a, mg-b), then issue the command.

Enabling Email Consolidation email consolidate 535-0008-00 Rev 10


217


Turns on email consolidation. The default is disabled.

Disabling Email Consolidation no email consolidate Turns off email consolidation.

Customizing Email Consolidation When email consolidation is enabled, the default settings are as follows:

• •

Time period before email is sent: 60 seconds Maximum number of events that can be combined: 5

Setting the Wait Period email consolidate period [time] Sets the wait time in seconds, before the email is sent.

Setting the Number of Combined Events email consolidate events [num] Sets the maximum number of combined events.

Viewing Current Email Settings You can view the current email consolidation settings with the following command. show email Output similar to the following is shown. vmem-acm-b [system1: master] # show email Mail hub: Mail hub port: 25 Domain: (system domain name) Return address: do-not-reply Include hostname in return address: yes ... Consolidate email settings: Enable: no Period: 60 Max_events: 5 ...

218


535-0008-00 Rev 10

Unified Logging Upload

Unified Logging Upload You can upload logs for all nodes within a Violin 6000 Series flash Memory Array (both Memory Arrays and internal Memory Gateways) with unified logging upload. A progress bar appears that tracks the upload across all nodes on which this procedure is performed. You can also perform log file uploads from each node separately.

Uploading Log Files for All Nodes You can use the logging files upload-auto command from the Violin 6000 Series internal Memory Gateway master (mg) to upload logs for all nodes within the Memory Array (both ACMs and both MGs). The logging upload-auto configuration on the internal Memory Gateway master is also the used for performing log file uploads for all other nodes.

Uploading Log Files for Specific Nodes You can perform log file uploads for each individual node of a Violin 6000 Series flash Memory Array (internal Memory Gateway master and standby, and ACM master and standby). In the case of a local upload, the logging upload-auto configuration on the local system is used. When you run the logging files upload-auto command from the ACM master, the uploaded log files are restricted to the ACMs (master and standby). In this case, log file uploads are not performed for the internal Memory Gateways (mg-a or mg-b).

Unified Logging Upload Commands This section covers the unified logging upload commands and their respective purposes.

Upload Log Files for All Nodes logging files upload-auto immediate When run from the internal Memory Gateway master, /var/log/messages are collected from both Memory Array nodes (ACM) and both internal Memory Gateway nodes (MG), and then uploaded to a pre-configured target.

Cancel a Log File Upload Process logging files upload-auto immediate cancel Stops an in-process log file upload.

Upload Log Files From the Local Host logging files upload-auto immediate local Only uploads /var/log/messages from the local host.

535-0008-00 Rev 10


219


Upload Log Files From a Specific Node logging files upload-auto immediate module id Uploads log files from a specified Node (acm-a/b, mg-a/b).

Upload Log Files From a Specific Node Type logging files upload-auto immediate module type Uploads log files from a specific node type (acm or mg).

220


535-0008-00 Rev 10

CHAPTER 6

Manage the Memory Array with the CLI

This chapter provides detailed procedures for configuring and showing status of the Memory Array and its internal modules. Most of these procedures can only be performed using the Command Line Interface (CLI). If the tasks in this chapter can also be performed using the Violin Web interface, references to those sections are provided. The following sections are included in this chapter:

• • • • • •

Overview of the Command Line Interface (CLI) on page 221 Connecting to the Command Line Interface (CLI) on page 227 Configuration File Management on page 230 Upgrading the System Software on page 233 Reboot and Shutdown Procedures on page 238 Rebalancing RAID Groups on page 242

See APPENDIX A, “CLI Reference” on page 287 for details on using the CLI commands.

Overview of the Command Line Interface (CLI) The CLI on the Memory Array is used to retrieve system status and set configuration information. The CLI can be used to perform all of the functions provided in the Violin Web Interface, as well as other functions that are not available in the Web Interface. Depending on the user’s current access mode (described in Command Modes on page 225), a subset of commands are available for configuring, managing and viewing information about the Violin 6000 Memory Array.

535-0008-00 Rev 10


221

Chapter 6: Manage the Memory Array with the CLI

CLI Shorthand Method Commands can be expressed in shorthand form in the CLI. Each keyword can be abbreviated by omitting its final letters, as long as the remaining letters are unique within the CLI command set. For example, the commands to display the system date and time or the hostname can be abbreviated as: sh clo

show clock

sh h

show hosts

Additional letters can be included but none can be skipped; for example, the show clock command can be typed as sho clo or sh clock or various other combinations, but not shw clk. Other commands that are frequently typed in shorthand include: en

enable

conf t

configure terminal

ex

exit

If the command is shortened too much, an error message appears and help is offered. For example, the abbreviation “sh cl” could mean show clock or show cluster so it generates this error message: > sh cl % Ambiguous command "cl". Type "sh cl?" for help.

Getting Help In any mode of the CLI you can query for help by using the help command or typing a question mark. See Command Modes on page 225 for descriptions of the three CLI modes that are available. The mode determines the set of commands that can be executed. Enter help at the prompt for a summary of how to use question marks to obtain context-sensitive help, as described here. Just entering a question mark ? by itself provides a list of available commands corresponding to the current mode. You can also query for options of a specific command by typing in the command, following it with a space, and adding a question mark. After displaying a list of options, the command line echoes the string and puts the cursor after it, ready for more input.

222


535-0008-00 Rev 10

Overview of the Command Line Interface (CLI)

For example, in Standard mode you can enter cli ? at the command line and see the following output. > cli ? clear-history

Clear the command history for the current user

session

Configure CLI options for this session only

> cli █ If the command is complete without further options or values, (carriage return) is displayed on a separate line and the command is echoed at the prompt. Pressing the Enter key (also known as carriage return, ) will then issue the command if no values are required, or will be displayed. For example: > cli session ? auto-logout Configure keyboard inactivity timeout for automatic logout paging Configure the ability to view text one screen at a time prefix-modes Configure the CLI's prefix modes feature for this session progress Configure progress updates for lon operations terminal et terminal parameters > cli session paging ? enable Enable paging > cli session paging enable ? > cli session paging enable > > cli session terminal ? length Set the number of lines for this terminal resize Resize the CLI terminal settings (to match with real terminal) type Set the terminal type width Set the width of this terminal in characters > cli session terminal width ?

535-0008-00 Rev 10


223


> cli session terminal width 60 > █ When is displayed and only specific values can be used (such as interface names or a VLAN identifier), those values will be displayed on new lines after . Similarly, when the command is complete but could include additional options, and the options are displayed, each on a separate line.

Tab Completion of Commands Commands can be completed by typing in the first few letters then pressing the Tab key (). Pressing Tab once completes the command if there is only one way to complete it; otherwise it expands the command to the next point of uncertainty. At that point, pressing Tab again displays a list of possible completions, which might be keywords or values, or both. h

Completes the help keyword.

sh

Completes the show keyword (but s does not, because more than one command starts with “s”).

show

Lists options of the show command that can immediately follow the show keyword.

sh

Completes the show keyword and lists options of the show command.

s

Lists all available commands that begin with “s”.

For a list of all commands currently available, press the Tab key twice at the prompt. In Standard mode, for example, press the Tab key twice to list these commands: >

224

cli

help

show

enable

no

slogin

exit

ping

telnet

terminal


535-0008-00 Rev 10

Overview of the Command Line Interface (CLI)

Command Modes The CLI can be in one of three modes, which determine the set of commands that can be executed. Commands that are not currently available do not show in help or completion, and generally behave as if they do not exist. Standard Mode

When the CLI is launched, it begins in Standard mode. This is the most restrictive mode and only has commands to query a restricted set of state information. In this mode you cannot take any actions that would directly affect the system, nor can you change any configuration. User accounts with the “unpriv” role are restricted to Standard mode. Enable Mode

The enable command moves the CLI to Enable mode. This mode has commands to view all state information and take certain kinds of actions, such as rebooting the system or configuring some system parameters, but it excludes commands that configure the cluster. Its commands are a superset of those in Standard mode. The disable command moves the CLI from Enable mode back to Standard mode. The exit command (in Enable mode) closes the CLI. User accounts with the “monitor” role can use all Enable mode commands. Config Mode

The configure terminal command moves the CLI from Enable mode to Config mode.

•

On the cluster’s master node, Config mode has a full unrestricted set of commands to view anything, take any action, or change any configuration. Its commands are a superset of those in Enable mode.

•

On nodes other than the master, Config mode only includes commands that operate on the local node. Using a global command on a standby or normal node either has a temporary local effect (which is overridden as soon as the node synchronizes with the master node) or produces an error message, which identifies the master node where the command can be used.

The exit command moves the CLI from Config mode to Enable mode. Using the exit command twice closes the CLI, or you can use the quit command to close the CLI directly. To return to Standard mode from Config mode, first exit Config mode and then disable the Enable mode. User accounts with the admin role can use all Config mode commands.

535-0008-00 Rev 10


225


Prompt and Response Conventions The prompt format is: [: ] The prompt begins with the hostname of the node and, in brackets, the cluster name and role of the node in that cluster (master or standby). The end of the prompt string indicates what command mode the CLI is in: > for standard mode, # for Enable mode, or (config) # for Config mode. For example, if the hostname of the master node is “gate1” and the cluster name is “vmgCluster” then the prompts for each of the CLI modes are: Standard mode:

gate1 [vmgCluster: master] >

Enable mode:

gate1 [vmgCluster: master] #

Config mode:

gate1 [vmgCluster: master] (config) #

The role can be master, standby, normal, or unknown. An asterisk (*) before the command prompt indicates that some configuration changes have not yet been saved to the active configuration file. For example, when changes need to be saved the command prompt for Config mode changes to this: Config (unsaved):

* gate1 [vmgCluster: master] (config) #

Configuration commands that succeed in doing what was asked do not print any response, so the next thing you see after pressing is another command prompt. You can verify the effect of a configuration command by using its corresponding show command to display current settings. If an error occurs in executing a command, the response begins with “%” followed by some text describing the error.

Abbreviations for Large Numbers The following abbreviations are used for large numbers in the output displays of various show and stats commands:

226

B

bytes

kB

kilobytes (10241 = 1,024 bytes)

MB

megabytes (10242 = 1,048,576 bytes)

GB

gigabytes (10243 = 1,073,741,824 bytes)


535-0008-00 Rev 10

Connecting to the Command Line Interface (CLI)

TB

terabytes (10244 = 1,099,511,627,776 bytes)

PB

petabytes (10245 = 1,125,899,906,842,624 bytes)

and so on for E (exabytes), Z (zettabytes), and Y (yottabytes). Single-letter abbreviations such as k, M, or G are sometimes used to conserve space, or for units other than bytes.

Connecting to the Command Line Interface (CLI) Logging into the Memory Array CLI via the Serial Interface Access the Command Line Interface (CLI) through console port A between the host computer and the Memory Array.

Connecting the Console Cable Use the serial console cable to connect a computer to the Memory Array for configuration operations and retrieval of system status using the command-line interface (CLI).

A B

1

A B

2

D

C

B

A

1

1

1

1

n D

n C

n B

n A

A

B

A B

ID A B

Console Port B

Console Port A

Figure 6.1 Connecting the Console Cable to the Memory Array

To connect the console cable to the Memory Array and a computer: 1. Connect one end of the console cable to serial console port A on the rear of the Memory

Array. 2. Connect the other end of the cable to a computer. 3. When you are ready to connect to the CLI, start a terminal application, and then use the

following serial console settings: — — — —

535-0008-00 Rev 10

Speed (baud): 115200 Data bits: 8 Stop bits: 1 Parity: None


227


4. Log in to the Master ACM as “admin”. The Master ACM is set up during initial configuration of

the Memory Array. Enter the system host name or the ACM management IP address to ensure a direct connection.

Logging in via a Windows Host Computer The following procedure shows how to connect to the Memory Array using a terminal emulation application. For example, using PuTTY to connect to the CLI through the serial connection: 1. Log into the host computer as an administrator. 2. From the Windows Start menu, select Programs > Accessories > Communications >

PuTTY. The New Connection - PuTTY1 window opens. 3. In the Connection Description window: a. Type the name of the new connection in the Name field. b. Select the appropriate icon from the list of available ones. c. Click OK when done.

The Connect To window opens. 4. In the Connect using drop-down list, select Serial. 5. Click Connect.

The following login prompt appears: violin login: 6. Type the default login name (admin) and password (admin) to log into the system.

You are now connected to the Master ACM of the Memory Array and can access the various CLI commands. Closing the PuTTY window terminates the connection.

Logging in to the CLI via a Linux host computer Connect to the Memory Array using the Kermit application or another terminal application, such as tip, if your system does not have Kermit. Save the Kermit parameter settings in a .kermrc file so that you do not need to type them each time you connect. The Violin Utilities package for Linux includes Violin’s version of Kermit called vtkermit.

Note: Place the

.kermrc file in your home directory so that the Kermit application recognizes it; otherwise, use the -y option followed by the location and file name of the configuration file. To connect to the CLI through the serial connection: 1. Log into the host computer as root. 2. At the host computer’s command line prompt, type: # kermit

228


535-0008-00 Rev 10

Connecting to the Command Line Interface (CLI)

or # vtkermit 3. Specify the line of the Kermit connection. At the Kermit prompt, type: C-Kermit> set line /dev/ttyS0

Note: The device name may vary depending on the host and the Linux operating system that is

running on the host. 4. Specify the flow control mode. At the Kermit prompt, type: C-Kermit> set flow xon/xoff 5. Specify the speed of the Kermit connection. At the Kermit prompt, type: C-Kermit> set speed 9600 6. Since the serial interface of the Memory Array does not require the carrier detect signal on the

Kermit connection, at the Kermit prompt, type: C-Kermit> set carrier-watch off 7. To establish the connection to the Memory Array, at the Kermit prompt, type: C-Kermit> connect

The following login prompt appears: violin login: 8. Type the default login name (admin) and password (admin) to complete logging into the

system. 9. At this point, you are connected to the CLI of the Memory Array and can access the various

CLI commands. To close the Kermit application, do one of the following: a. At the CLI prompt, type Ctrl-\, c, and then q. b. Close the terminal window, which terminates the Kermit connection.

Logging in to the CLI Using an SSH Client You can connect to the Memory Array CLI using an SSH client. The following procedure is an example of connecting to the Memory Array using PuTTY. 1. Start the SSH program. 2. In the Category > Session window, type either the host name of the Memory Array Master

ACM or its IP address in the Host Name (or IP address) field. 3. Set the Connection type to SSH. 4. Click the Open button at the bottom of the screen. 5. Enter the admin password to log in to the Memory Array.

535-0008-00 Rev 10


229


Configuration File Management Changes made to the configuration of a cluster take effect immediately, but those changes can be lost if they are not saved to a configuration file. The Memory Array stores one or more configuration files on persistent storage, one of which is designated as active. The active configuration file is where changes are stored when you save the configuration. Whenever the system reboots, it loads the configuration settings from the file designated as active. The system backs up the active configuration file automatically.

Saving the Configuration You can save the configuration by clicking on the Commit Changes button in any screen of the Web Interface, or by using one of these commands (which have the same effect) in the CLI:

• •

configuration write write memory

An asterisk (*) before the command prompt is a reminder that changes were made but have not yet been saved; therefore after saving the configuration, the asterisk no longer appears. Example of configuration write command:

* violin-acma [violin: master] (config) # configuration write violin-acma [violin: master] (config) # █

Example of write memory command:

* violin-acma [violin: master] (config) # write memory violin-acma [violin: master] (config) # █

Either command (or the Commit Changes button) stores all current configuration parameters to persistent storage in the active configuration file.

Web: In the Web Interface, the Commit Changes button appears in the message bar only when the current configuration includes changes that have not yet been saved; otherwise the message bar displays a “No unsaved changes” message. You can switch to another configuration file by saving the current configuration to a specified filename.

• •

In the Web Interface, use Save Configuration as: command.

•

In the CLI, use configuration write to command.

This option is at the bottom of the Cluster Administration screen. First enter a filename, then click the SAVE button.

For example, the following command saves the configuration to a file named cf1 and makes it the active configuration file:

230


535-0008-00 Rev 10

Configuration File Management

violin-acma [violin: master] (config) # configuration write to cf1 violin-acma [violin: master] (config) # █

If you want to save the current configuration to a different file without making it active, use the noswitch option of the configuration write to command. This option can be used to save an interim configuration which you will finish later. Saving with the no-switch option does not change a configuration’s unsaved status (indicated by an asterisk in the CLI), because recent changes are still not stored in the active configuration file. For example:

* violin-acma [violin: master] (config) # configuration write to cf2 no-switch * violin-acma [violin: master] (config) # █

Showing Configuration Files You can display a list of configuration files in the Cluster Administration screen (see Cluster Admin on page 114), or by using this CLI command:

•

show configuration files

The display shows all configuration files associated with this system and identifies the active file, which is backed up as .bak. Backups of previously active configuration files also appear in the display. For example, after switching from the original configuration file (initial) to cf1 and saving to cf2 with the no-switch option, as shown above, the display lists these five configuration files:

violin-acma [violin: master] (config) # show configuration files initial initial.bak cf1 (active) cf1.bak cf2 violin-acma [violin: master] (config) # █

Showing Configuration The CLI commands that would bring the state of a fresh system up to match a given configuration can be displayed by using one of the following commands:

•

show configuration - display commands to reproduce the state saved in the active configuration file.

•

show configuration running - display commands to reproduce the state of the current running configuration, which could include some unsaved configuration settings.

• • •

show running-config - same as show configuration running. write terminal - same as show configuration running. show configuration files - display commands to reproduce the state saved in the specified configuration file.

535-0008-00 Rev 10


231


•

show configuration full - display all commands, including any default settings, to reproduce the state saved in the active configuration file.

•

show configuration running full - display all commands, including any default settings, to reproduce the state of the current running configuration.

The output of each of these commands begins with a short header that contains the name and version number of the configuration, in a comment. Except for the show configuration full and show configuration running full commands, the output omits commands that are not required because they set parameters to their default values.

Reverting without Saving If you do not want to keep the current changes, you can revert to the last saved configuration or apply an inactive configuration file to the system. (Applying the active configuration file has the same effect as reverting without saving.) In the CLI, you can use the configuration revert saved command to undo recent changes and return to the configuration saved in the active file. For example:

violin-acma [violin: master] (config) # configuration revert saved violin-acma [violin: master] (config) # █

Web: In the Violin Web Interface, you can select any available configuration file in the Cluster Administration screen and use the APPLY button to apply its configuration to the system.

Deleting, Moving, and Copying Configuration Files You can delete, move (rename), or copy an inactive configuration file by using these CLI commands:

• • •

configuration delete configuration move configuration copy

An active configuration file may not be deleted or renamed, nor may it be the target of a move or copy. It may be the source of a copy, in which case the original remains active.

Web: In the Current Configurations section of the Cluster Administration screen, you can select one or more configuration files to delete. First click the check box for each file you want to delete, then use the DELETE button.

232


535-0008-00 Rev 10

Upgrading the System Software

Upgrading the System Software Note: This section is specific to upgrading from release V6.0.0 to later releases. If you want to upgrade a system running V5.x.x, see the V6.0.0 Release Notes or contact Violin Memory Customer Support, as described in Contacting Violin Memory on page 5.

This section provides the basic steps required to upgrade the Memory Array firmware and Memory Gateway software using the CLI.

•

Before attempting to upgrade the system, always review the latest Release Notes for the most up-to-date instructions, information on the necessary software image and required upgrade paths.

•

To upgrade Memory Array firmware or Memory Gateway software using the Web interface, see Upgrading the Violin Cluster on page 138.

Downloading the Upgrade Image Follow these steps if you will be downloading the image(s) to a USB drive.

Note: The USB drive must be FAT32 formatted.

Note: You must register as a Violin Memory customer to download software updates. If not

already registered, go to http://www.violin-memory.com/support/register 1. Go to http://www.violin-memory.com/support/ 2. Log in to Customer Support using your Violin Memory Customer Portal login and password

information. (Click Register if you do not have an account.) 3. Click the Software Downloads tab. 4. From the Violin Memory Array 6000 Series location, download the appropriate image.

Upgrading All Firmware and Software This section provides the steps required to upgrade both the Memory Array firmware and the Memory Gateway software together. See the following sections if you want to update the software and firmware separately:

• •

Upgrading the Memory Gateway Software on page 235 Upgrading the Memory Array Firmware on page 236

WARNING! Make sure that all alarms are cleared before starting an upgrade process.

535-0008-00 Rev 10


233


To upgrade the system: Skip to step 3 if you are not upgrading the image from a USB drive. 1. If upgrading from a USB drive, make sure that the image is loaded on the USB drive. 2. Insert the USB drive into the USB port on the front of the Memory Array that is associated with

the Master Memory Gateway node: the left USB port is for MG A; the right USB port is for MG B. 3. Start a terminal application and log in to the Master Memory Gateway as admin. 4. At the command prompt, type enable and then configure terminal:

violin-mga [violin: master] > enable violin-mga [violin: master] # configure terminal█ 5. Perform an upgrade using one of the following methods, depending on the location of the

image file: a. To upgrade the image from a USB drive, type the following commands to mount the

USB drive and then install the software image: (config) # image fetch usb (config) # system upgrade all immediate█ b. To upgrade the image from a URL accessible to the system, type the following, where

is the path to the image: (config) # system upgrade all immediate█ This upgrades the Memory Gateway software and Memory Array firmware simultaneously. During this process clients are unable to maintain connections to exported LUNs until the system is restarted. Although this method of upgrading the system is relatively quick, you experience some downtime while the system reboots.

Note: The upgrade process may take from 20 minutes to 1 hour to complete. Do not power off

or reboot the system during the upgrade process.

Once all modules are updated with the new software, both Memory Gateways shut down. The ACMs then reboot. The VCMs and VIMMs are updated, if necessary. The Memory Gateways then boot up with the new version of software. To monitor the upgrade process: Wait 1-2 minutes for the system to reboot and then log in to the Master ACM to monitor the upgrade process, if desired.

234

•

Type monitor upgrades to monitor the VCM upgrade process.

•

Once the VCM upgrade process is completed, type vdiag to show the VIMM upgrade progress.


535-0008-00 Rev 10


Upgrading the Memory Gateway Software This section provides the steps required to only upgrade the Memory Gateway software. If a firmware update is also available for the Memory Array, it is recommended that you upgrade the Memory Array firmware and the Memory Gateway software together. This process is described in Upgrading All Firmware and Software on page 233.

Note: Only the Memory Gateway software is upgraded using these instructions. Ensure that the

Memory Array is running compatible versions of Memory Gateway software and Memory Array firmware. See the latest Release Notes or contact Violin Memory Customer Support for more information.

There are two methods available for upgrading the Memory Gateways: staged (non-disruptive) and immediate. Both are described in this section. Skip to step 3 if you are not upgrading the software from a USB drive. 1. If upgrading from a USB drive, make sure that the software image is loaded on the USB drive. 2. Insert the USB drive into the USB port on the front of the Memory Array that is associated with

the Master Memory Gateway node: the left USB port is for MG A; the right USB port is for MG B. 3. Start a terminal application and log in to the Master Memory Gateway as admin. 4. At the command prompt, type enable and then configure terminal:

violin-mga [violin: master] > enable violin-mga [violin: master] # configure terminal█ 5. Perform an upgrade using one of the following methods, depending on the location of the

software image file: a. To upgrade the software from a USB drive, type the following commands to mount the

USB drive and then install the software image: (config) # image fetch usb (config) # system upgrade mg staged|immediate█ b. To upgrade the software from a URL accessible to the system, type the following,

where is the path to the image: (config) # system upgrade mg staged|immediate█

•

For a staged upgrade, the cluster is split into two clusters, A and B, with one Memory Gateway in cluster A and the other in Cluster B. The Memory Gateway in Cluster B is upgraded first, followed by Cluster A. While Cluster B is being upgraded, Cluster A provides access to storage, and vice-versa.

•

For an immediate upgrade, the Memory Gateways are upgraded simultaneously. During this process clients are unable to maintain connections to exported LUNs until the cluster is restarted. Although this method of upgrading the cluster is relatively quick, you experience some downtime while the nodes reboot.

Both Memory Gateways will upgrade and then reboot on their own. Once the Gateways reboot, the command prompt returns to the ACM Master.

535-0008-00 Rev 10


235


6. Wait 1-2 minutes and then type “show array modules type mg detail” to see if the

Gateways have booted. The Gateways are powered on if their IP addresses are shown in the output.

Upgrading the Memory Array Firmware If you are only upgrading the Memory Array firmware, the Memory Gateways must be shut down and powered off before upgrading the Memory Array. If a software update is also available for the Memory Gateways, it is recommended that you upgrade the Memory Array firmware and the Memory Gateway software together. This eliminates the need to manually shut down the Memory Gateways. This process is described in Upgrading All Firmware and Software on page 233.

Shutting Down the Memory Gateways 1. Start a terminal application and log in to the Master ACM as “admin”. The Master ACM is set

up during initial configuration of the Memory Array. Enter the system host name or the ACM management IP address to ensure a direct connection. 2. At the command prompt, type enable and then configure terminal:

violin-acma [violin: master] > enable violin-acma [violin: master] # configure terminal█ 3. Log in to the Master Memory Gateway by typing the following command:

violin-acma [violin: master] # slogin mg-master█ 4. At the command prompt, type enable and then configure terminal:

violin-mga [violin: master] > enable violin-mga [violin: master] # configure terminal█ 5. Shut down the internal Memory Gateways by typing the following command:

violin-mga [violin: master] (config) # cluster shutdown█ 6. Respond “yes” to the shutdown confirmation question.

The command prompt returns to the ACM Master once the Memory Gateways shut down. 7. From the ACM Master command prompt, type the following to verify that the Memory

Gateways have shut down: (config) # show array modules type mg detail█ The Memory Gateways are shut down when the "Running" status is "no". 8. Power down the Memory Gateways by typing the following command:

(config) # no array modules type mg enable WARNING: Powering down array module(s) may cause data disruption. Are you sure? [no]█ Note: It will take from 30 seconds to 3 minutes for the Memory Gateways to shut down.

236


535-0008-00 Rev 10


The Memory Gateways power on automatically after the Memory Array firmware is upgraded. To upgrade the Memory Array firmware:

Note: Make sure the Memory Gateways are shut down and powered off before continuing. 1. If upgrading from a USB drive, make sure the new image file is loaded on the USB drive, as

described in Downloading the Upgrade Image on page 233. There are two USB ports on the Memory Array front panel, one for each ACM. Only the USB port associated with the current Master node can be used to upgrade the system. 2. Start a terminal application and log in to the Master ACM as “admin”. The Master ACM is set

up during initial configuration of the Memory Array. Enter the system host name or the ACM management IP address to ensure a direct connection. 3. Log in to the Memory Array as admin. 4. At the command prompt, type enable and then configure terminal:

violin-acma [violin: master] > enable violin-acma [violin: master] # configure terminal█ 5. Identify which ACM is the master by typing the following command:

violin-acma [violin: master] (config) # show cluster master Node Status: Node ID: 1 Host ID: 9075d6177c48 Hostname: qa-stein16-acmb Node Role: master Node State: online Node internal address: 169.254.1.11, port: 60102 Node external address: 10.1.14.208 Recv. Heartbeats from: 2 Send Heartbeats to: 2 6. If upgrading from a USB drive, insert the USB drive into the USB port on the front of the

Memory Array that is associated with the Master node: the left USB port is ACM A; the right USB port is ACM B. The LED next to the USB port will start flashing. 7. Perform an upgrade using one of the following methods, depending on the image location: a. To upgrade from a USB drive, type the following to mount the USB drive and then install

the image: (config) # usb mount (config) # cluster upgrade usb:// immediate█ b. To upgrade from a URL accessible to the system, type the following, where is

the path to the image: (config) # cluster upgrade immediate█ This upgrades both ACMs and reboots both modules immediately after completion of the upgrade.

535-0008-00 Rev 10


237


8. If upgrading from a USB drive, log back in to the same ACM from which the upgrade was done

and then type the following at the command prompt to eject the USB drive: violin-acma [violin: master] (config) # usb eject█ Note: The upgrade process may take from 20 minutes to 1 hour to complete. Do not power off or reboot the system during the upgrade process. Type monitor upgrades at the command prompt to monitor the VCM upgrade process. Once the VCM upgrade process is completed, type vdiag to show the VIMM upgrade progress. The Memory Gateways power on when the VCMs are ready.

Reboot and Shutdown Procedures This section provides instructions for several different types of reboot and shutdown procedures. To ensure proper system functionality, complete each step in the order presented.

Before You Proceed It is recommended that you complete the following before rebooting or shutting down the Memory Array or any of its modules.

Save the Configuration File Changes made to a cluster can be lost if they are not saved to a configuration file. The Memory Array stores one or more configuration files on persistent storage, one of which is designated as active. The active configuration file is where changes are stored when you save the configuration. Whenever the system reboots, it loads the configuration settings from the file designated as active. The system backs up the active configuration file automatically. To save changes made to the configuration file: You can save the configuration by clicking the Commit Changes button in any screen of the Web Interface, or by using one of these commands (which have the same effect) in the CLI: * violin-acma [violin: master] (config) # configuration write█ violin-acma [violin: master] (config) # write memory█

An asterisk (*) before the command prompt is a reminder that changes were made but have not yet been saved; therefore after saving the configuration, the asterisk no longer appears. See Configuration File Management on page 230 for more information.

238


535-0008-00 Rev 10

Reboot and Shutdown Procedures

Rebooting the Memory Array The following command reboots the Memory Array and all of its modules. See Array and Cluster Action Commands on page 313 for the commands to reboot all Array Controller Modules, Memory Gateways or vRAID Controller Modules. 1. Review the procedures described in Before You Proceed on page 238. 2. log in to the Memory Array as admin. 3. At the command prompt, type enable and then configure terminal:

violin-acma [violin: master] > enable█ violin-acma [violin: master] # configure terminal█ 4. Shut down the Memory Gateways as described in Shutting Down the Memory Gateways on

page 236. 5. Once the Memory Gateways are shut down, return to the Master ACM and then type the

following at the command prompt (from config mode): violin-acma [violin: master] (config) # array reboot WARNING: Rebooting array may cause data disruption. Are you sure? [no]█ 6. Type “yes” to verify the command.

The following modules are rebooted: Array Controller Modules, vRAID Controller Modules, internal Memory Gateways, VIMMs and Host Bus Adapters. Depending on the number (24 or 64) and type of VIMMs (SLC or MLC), the reboot process will take at least one hour for an SLC system and as long as two hours for an MLC system. Type the following command to view reboot progress: violin-acma [violin: master] (config) # monitor boot█

Shutting Down the Memory Array Follow these steps to shut down all Memory Array modules. 1. Review the procedures described in Before You Proceed on page 238. 2. log in to the Memory Array as admin. 3. At the command prompt, type enable and then configure terminal:

violin-acma [violin: master] > enable█ violin-acma [violin: master] # configure terminal█ 4. Shut down the Memory Gateways as described in Shutting Down the Memory Gateways on

page 236. 5. Once the Memory Gateways are shut down, return to the Master ACM and then type the

following at the command prompt (from config mode): violin-acma [violin: master] (config) # array shutdown█

535-0008-00 Rev 10


239


Powering Off and On Memory Array Modules Follow these steps to power off and power on all Memory Array module types. 1. Log in to the Memory Array as admin. 2. At the command prompt, type enable and then configure terminal:

violin-acma [violin: master] > enable█ violin-acma [violin: master] # configure terminal█ 3. Power off or power on modules, as follows:

•

To power off all types of a Memory Array module, type the following at the command prompt:

violin-acma [violin: master] (config) # no array modules type enable█ Where:

•

vcm

is the vRAID Controller Modules in the Memory Array. This powers off all VCMs and VIMMs.

acm

is the Array Controller Modules in the Memory Array.

fpm

is the Front Panel Module in the Memory Array.

mg

is the internal Memory Gateways in the Memory Array. This powers off all Memory Gateways.

hba

is the Host Bus Adapters in the Memory Array. This powers off all HBAs.

To power on a specific type of Memory Array module, type the following at the command prompt:

violin-acma [violin: master] (config) # array modules type enable█

Powering Off and On Specific Memory Array Modules The following Memory Array modules can be powered off individually: internal Memory Gateways, Host Bus Adapters, VIMMs and VCMs. Type the following at the command prompt to power off a specific module: violin-acma [violin: master] (config) # no array modules id enable█ Where:

240

vimmxx

xx the slot number of the VIMM (0-63) in a 64-VIMM system

vcm-x

x is one of the VCMs (vcm-a, vcm-b, vcm-c, vcm-d)

mg-x

x is one of the Memory Gateways (either mg-a or mg-b)

hba-x

x is one of the Host Bus Adapters (hba-a, hba-b, hba-c or hba-d)


535-0008-00 Rev 10

Reboot and Shutdown Procedures

Type the following at the command prompt to power on a specific module: violin-acma [violin: master] (config) # array modules id enable█ Some modules must be powered off and on when being replaced. See the Violin 6000 Series Memory Array Service Guide for instructions on replacing Memory Array modules.

Restoring Power to VIMMs and VCMs VIMMs and VCMs that are powered off or that have failed are flagged by the system as a module that should not be used or may need to be replaced. This flag—which can be displayed in the CLI by typing “show donotuse”—is a reminder to the system administrator if a VIMM or VCM fails. This flag may also display for VIMMs and VCMs that do not require replacement. The following is an example alarm for a VIMM that has the “donotuse” flag: # VIMM 1 faulted; VIMM is not responding; VIMM taken out of service█ To clear the donotuse flag: Before you can power on a VIMM or VCM, you must clear the donotuse flag if it is on. 1. First, check to see if the flag is on by typing the “show donotuse” command at the prompt. If

the flag has been set for a VIMM by the system, you will see output similar to the following: # show donotuse vimmid system VIMM is not responding at 11-01-2012 20:41:11 UTC█ If the administrator set the flag, by issuing the no array modules id vimmid command, the output would be: vimmid user-set VIMM is not responding at 11-01-2012 20:41:11 UTC 2. Type one of the following at the command prompt: a. To clear the flag for a VIMM, where id is the VIMM number:

# no array modules id vimmid donotuse█ b. To clear the flag for a VCM, where x is the VCM letter (a, b, c or d):

# no array modules id vcm-x donotuse█ c. To clear the flag for all system-set faults, where component is either VIMM or VCM:

# no array modules id donotuse█ 3. See Powering Off and On Specific Memory Array Modules on page 240 to complete the

power-on sequence. Note: If the VIMM or VCM has failed and requires replacement, see the Violin 6000 Series

Memory Array Service Guide for instructions on replacing Memory Array modules.

535-0008-00 Rev 10


241


Restarting the Memory Array Follow these procedures to restart the Memory Array if it has been shut down.

Caution: The Memory Array must be completely shut down before proceeding. 1. Unplug the AC power cords from the power receptacles on the rear of the Memory Array

chassis. 2. Wait 30 seconds and then reconnect the power cords to their power receptacles.

The Memory Array begins to start up. During the boot up process, the green Status LEDs will flash. The Memory Array is fully booted when the Status LED is continuously lit.

Rebalancing RAID Groups If RAID groups become unbalanced over time due to VIMM replacement or other administratorinitiated changes, the VIMM tree can be manually or automatically balanced as needed. An administrator can also schedule automatic Array balancing. Unbalanced RAID groups are reported in the Violin Web interface. Auto balance is disabled by default. If two or more VIMMs in a RAID group share the same root VIMM, the RAID group is considered “unbalanced”, causing the system to perform less than optimally. Balance of the system can be restored by “moving” VIMMs from RAID rebuilds. If necessary, multiple RAID rebuilds can be performed at the same time. Unprotected RAID groups cannot be rebalanced until the parity VIMM is fully recovered and a spare VIMM is available. Each RAID group consists of five VIMMs: four for data an one parity. The parity VIMM is used to reconstruct data. A RAID rebalance can be set to occur automatically, on a scheduled basis or manually. A RAID rebuild can take three hours or longer. Table 6.1 shows all of the automatic, scheduling and manual commands. Examples of these commands are described in the sections that follow.

Note: If one of the balance functions is enabled (automatic or scheduled), it must be disabled

before setting up a different balance function.

Command

Description

array balance

This starts a RAID rebuild for each VCM that requires a rebuild. More than one rebuild may be required to balance the system.

array balance enable

This command, which is the default setting, enables automatic balancing. When unbalanced RAID groups are detected, they are automatically rebuilt.

Table 6.1 Array Rebalance Commands

242


535-0008-00 Rev 10

Rebalancing RAID Groups

Command

Description

no array balance enable

Use this command to disable (turn off) automatic rebalancing.

show array balance

Display the current RAID rebalance settings and view the status of all the VCMs.

array balance schedule once

This command schedules an array balance for the default time of the upcoming Saturday at 1:00 A.M. This starts a RAID rebuild for each VCM that requires a rebuild. More than one rebuild may be required to balance the system.

array balance schedule once [day-of-week ] [time ]

Use this command and parameters to set a one-time RAID rebalance at a specified time and day of the week.

array balance schedule weekly [day-of-week ] [time ]

Use this command and parameters to set a specific time and day of the week on a recurring basis. Multiple RAID rebuilds are completed until the system is balanced. The scheduled array balance is disabled once the system is in balance.

no array balance schedule once

Cancel a scheduled one-time rebalance.

no array balance schedule weekly Cancel scheduled weekly recurring rebalances. Table 6.1 Array Rebalance Commands

Discovering Unbalanced RAID Groups The System Status LED on the front of the Memory Array is amber when an alarm has been raised on at least one of the following modules: VIMM, MG, HBA, VCM, ACM, PCM, Fan, Power Supply. Use the “show alarms” command to identify alarms. Also, issue the “show array balance” command (see page 245) or the “vdiag” command to identify RAID groups that require rebalancing. Following is an example of vdiag: violin-acma [violin: master] > enable violin-acma [violin: master] # vdiag ... --- reading VIMM Config for all VCMs - VIMM Config for VCM: vcm-a... Degraded *** WARNING: Unbalanced VIMMs - 6 & 1 have the same root port : 1 - VIMM Config for VCM: vcm-b... Optimal - VIMM Config for VCM: vcm-c... Degraded *** WARNING: Unbalanced VIMMs - 6 & 1 have the same root port : 1 *** WARNING: Unprotected Raid Group(s): 0,1 - VIMM Config for VCM: vcm-d... Optimal ... The Cluster Status page in the Web interface will also display a message in yellow if any RAID group is unbalanced. Click Home > System View to view the Cluster Status page.

535-0008-00 Rev 10


243


Enabling Automatic Rebalancing By setting up automatic rebalancing, VCMs will initiate RAID rebuilds automatically whenever necessary. The setting is disabled once all VCMs report that the system is balanced. 1. Start a terminal application and log in to the Master ACM as “admin”. The Master ACM is set


violin-acma [violin: master] > enable violin-acma [violin: master] # configure terminal█ 3. Type the following command:

(config) # array balance enable█ Unbalanced RAID groups are automatically rebuilt when detected. The following is an example of a RAID rebuild in progress: RAID rebuild due to rebalance in progress, group 1, VIMM 49 (0% complete) To turn off automatic rebalancing: Use the following command to turn off automatic rebalancing: # no array balance enable█

Enabling Incremental Rebalance of RAID Groups Follow these steps to set up incremental balancing, which starts a rebuild of any unbalanced RAID groups now. If any RAID groups are unbalanced, only one rebuild is performed to restore balance. 1. Start a terminal application and log in to the Master ACM as “admin”. The Master ACM is set


violin-acma [violin: master] > enable violin-acma [violin: master] # configure terminal█ 3. Type the following command to rebuild any unbalanced RAID groups now:

(config) # array balance█ Each VCM is checked for any needed RAID rebalance. If necessary, a RAID rebuild is then immediately started.

244


535-0008-00 Rev 10

Rebalancing RAID Groups

Scheduling a RAID Group Rebalance Scheduling regular RAID rebalancing may be a more convenient option so you don’t have to complete the occasional manual check for unbalanced RAID groups. 1. Start a terminal application and log in to the Master ACM as “admin”. The Master ACM is set


violin-acma [violin: master] > enable violin-acma [violin: master] # configure terminal█ 3. Enter one of the following commands: a. To schedule a one-time RAID rebalance at the default day and time of 1:00 A.M. Saturday:

# array balance schedule once█ b. To schedule a one-time RAID rebalance at a specific time and day of the week:

# array balance schedule once [day-of-week ] [time ]█ c. To schedule a recurring weekly RAID rebalance:

# array balance schedule weekly [day-of-week ] [time ]█ To turn off a scheduled RAID group rebalance: Use the following command to turn off a scheduled RAID rebalance: # no array balance schedule █

Showing Current RAID Rebalance Settings Follow these steps to display the current RAID rebalance settings and to view the status of all the VCMs. 1. Start a terminal application and log in to the Master ACM as “admin”. The Master ACM is set

up during initial configuration of the Memory Array. Enter the system host name or the ACM management IP address to ensure a direct connection. 2. Type the following at the command prompt:

violin-acma [violin: master] > enable violin-acma [violin: master] # show array balance auto balance : no one-time balance : day-of-week: sat, time: 01:00:00 (next: 2012/ 11/24 01:00:00 -0800) weekly balance : no vcm-a vcm-b vcm-c vcm-d

535-0008-00 Rev 10

: : : :

balance in progress balanced balanced unbalanced


245


246


535-0008-00 Rev 10

CHAPTER 7

Data Security

This chapter covers available options for enhancing data security and includes the following topics:

• • • • •

Managing Data Security on page 247 iSCSI Authenticated Sign-On on page 248 Setting Up iSCSI CHAP Authentication on page 248 Removing CHAP Secrets from a Target on page 251 Showing iSCSI CHAP Configurations on page 252

Managing Data Security Managing data security is the practice of ensuring data is protected from destructive forces and the intrusive actions of unauthorized users. Violin Memory ensures the security of your data with the following features:

•

iSCSI CHAP—Enforces iSCSI Authenticated Sign-On to ensure that Internet Small Computer System Interface (iSCSI) initiators and targets prove their identity to each other using the Challenge-Handshake Authentication Protocol (CHAP). Note:

535-0008-00 Rev 10

iSCSI CHAP Authentication requires the use of an iSCSI target.


247

Chapter 7 - Data Security

iSCSI Authenticated Sign-On The Challenge-Handshake Authentication Protocol (CHAP) authenticates a user or network host, validating the identity of the remote clients before allowing them access to an array. Using CHAP within iSCSI enhances security through the prevention of dictionary attacks, spoofing, and reflection attacks. There are two types of CHAP authentication to choose from: Unidirectional CHAP Authentication and Bidirectional CHAP Authentication.

Unidirectional CHAP Authentication Unidirectional CHAP is also known as One-Way CHAP. Unidirectional CHAP allows the target to verify the credentials of the initiator before allowing it to proceed to the login process. Once the authentication negotiation is completed, the target generates a challenge phrase using its own CHAP secret words and ID, then sends it to the initiator. The initiator uses the known target CHAP secret to create a response, and sends the response to the target. After the initiator’s response is successfully verified by the target, the login process begins.

Bidirectional CHAP Authentication Bidirectional CHAP is also known as Two-Way CHAP, or Mutual Authentication. Bidirectional CHAP allows an initiator to verify that it is attempting to log in to the correct target. This option is dependent on the client set up (e.g., iSCSI initiator). After the initiator has been successfully authenticated by the target, the initiator creates a CHAP challenge phrase that it sends to the target. In this instance, the target uses the known initiator CHAP secret to generate a response. After the target’s response is successfully verified, the initiator will continue with the login process.

Setting Up iSCSI CHAP Authentication Review CHAP Requirements on page 249 for CHAP secret recommendations and restrictions, then choose one of the following configuration methods.

• • •

Establishing a CHAP Secret on page 249 Configuring Unidirectional iSCSI CHAP on page 250 Configuring Bidirectional iSCSI CHAP on page 250

Note: When you enable CHAP on a Memory Gateway (target) for one client (initiator), the rest of the clients (for that target) must also use CHAP. Clients have to enable CHAP (from their side) before they are able to log in to the target.

248


535-0008-00 Rev 10

Setting Up iSCSI CHAP Authentication

CHAP Requirements iSCSI CHAP configurations must meet the following requirements:

• • •

The minimum CHAP secret size is 12 (bytes) characters (IPSec is not supported).

•

For bidirectional CHAP, each target-initiator pair must have one target CHAP secret and one initiator CHAP secret configured.

•

In bidirectional CHAP mode, the target CHAP secret cannot be identical to the initiator’s CHAP secret to which it is connected.

•

A particular target can have bidirectional CHAP set up with multiple clients.

The maximum CHAP secret size is 64 (bytes) characters. For unidirectional CHAP, each target-initiator pair must have one target CHAP secret configured.

Note: iSCSI CHAP configurations are persistent across system reboots.

Establishing a CHAP Secret This section demonstrates the ways in which you can establish a CHAP secret. For information on the complete iSCSI CHAP configuration process, see Configuring Unidirectional iSCSI CHAP on page 250 or Configuring Bidirectional iSCSI CHAP on page 250.

Note: A CHAP secret must have a minimum of 12 characters (bytes). The maximum length is 64

characters.

•

Use the CLI command {target-secret | initiator-secret} , for example: iscsi chap target iscsi target-secret Myfirstsecret to initiator iqn.1991-05.com.microsoft:violin-w2k3r2

•

Enter the secret at the prompt at the “CHAP secret” prompt using {target-secret-prompt | initiator-secret-prompt} format. In this case the character input is hidden, as shown in the following example: iscsi chap target iscsi initiator-secret-prompt to initiator iqn.1991-05.com.microsoft:violin-w2k3r2 CHAP secret: ***************

535-0008-00 Rev 10


249

Chapter 7 - Data Security

Configuring Unidirectional iSCSI CHAP In a unidirectional iSCSI CHAP configuration, the Memory Gateway (target) authenticates the client (initiator) to make sure it is a trusted host. This configuration requires the establishment of a target CHAP secret. To configure unidirectional CHAP authentication, do the following: 1. Connect to the Memory Gateway master through a terminal port. 2. Enable configuration mode, using the following commands.

> enable # configure terminal (config) # 3. Establish the target CHAP secret for a specified initiator using the following command,

substituting the variables with information for your site. See CHAP Requirements on page 249 for CHAP secret recommendations and restrictions. # iscsi chap target {target-secret|target-secretprompt} to {|}

Note: The igroup_name variable option, as used in this command, allows you to establish the CHAP secret for a group of initiators all at once. However, when you view iSCSI CHAP configurations, as described in Showing iSCSI CHAP Configurations on page 252, the individual initiators that are members of the igroup are shown individually without reference to the igroup_name (they belong to).

Configuring Bidirectional iSCSI CHAP In a bidirectional iSCSI CHAP configuration, the Memory Gateway (target) authenticates the client (initiator) to make sure it is a trusted host. Then, the client authenticates the target to ensure that it is a trusted device. This configuration requires the establishment of a target CHAP secret and an initiator CHAP secret. iscsi chap target {target-secret|target-secretprompt|initiator-secret|initiator-secret-prompt} to initiator {|igroup } To configure bidirectional CHAP authentication, do the following: 1. Connect to the Memory Gateway master through a terminal port. 2. Enable configuration mode, using the following commands.

> enable # configure terminal (config) # 3. Establish a target (Memory Gateway) CHAP secret, as described in Configuring Unidirectional

iSCSI CHAP on page 250.

250


535-0008-00 Rev 10

Removing CHAP Secrets from a Target

4. Establish the initiator CHAP secret using the following command. Substitute target variables

with the values used to establish the target CHAP secret, and initiator variables with the appropriate initiator information. See CHAP Requirements on page 249 for CHAP secret recommendations and restrictions. # iscsi chap target {initiator-secret|initiatorsecret-prompt} to {|}

Note: The igroup_name variable option, as used in this command, allows you to establish the CHAP secret for a group of initiators all at once. However, when you view iSCSI CHAP configurations, as described in Showing iSCSI CHAP Configurations on page 252, the individual initiators that are members of the igroup are shown individually without reference to the igroup_name (they belong to).

Removing CHAP Secrets from a Target Replacing of a CHAP secret of an existing target-initiator is not supported. Please delete the old CHAP secret and enter a new one. The igroup here is used like an expander that helps to set up the CHAP secret for a group of initiators in one CLI command. However it doesn't have the same meaning as used in the lun export CLI command. To remove the target secret and initiator secret for a target-initiator pair, do the following: 1. Connect to the Violin Memory Gateway master through a terminal port. 2. Enable configuration mode, using the following commands.

> enable # configure terminal (config) # 3. Remove the initiator CHAP secret for a target and initiator pair using the following command.

Substitute target and initiator variables with the values appropriate for each device. # no iscsi chap target to {| enable # configure terminal

Creating a Thin Provision LUN The command for creating a thin provision LUN provides the same options as the command for creating a regular LUN, with the exception of thin at the end of the syntax. # lun create container name size [ | equal] [quantity ] [nozero] [readonly] [startnum ] [blocksize 512 | 4096] [offline] thin

535-0008-00 Rev 10


261

Chapter 8 - Data Management

The following example demonstrates the creation of a 10 GB thin provision LUN named Mylun1 in a container named “MyContainer”. # lun create container MyContainer name Mylun1 size 10 thin LUN create: success!

Modifying Read/Write Access for a Thin Provision LUN The lun set command allows you to specify a LUN as read-only. This command is the same as that used to modify the read/write access for a regular (thick) LUN. # lun set container name readonly

Renaming a Thin Provision LUN You can use the lun rename command to rename a thin provisioned LUN. # lun rename container name to

Exporting a Thin Provision LUN The process for exporting thin provision LUNs is the same as exporting regular (thick) LUNs. For more information, see Exporting LUNs on page 193.

Note: Once a LUN is exported, it cannot be deleted without first unexporting the LUN.

# lun export container name [lunid][to ]... [using ]

Unexporting a Thin Provision LUN The process for unexporting thin provision LUNs is the same as exporting regular (thick) LUNs. For more information, see Exporting LUNs on page 193.

Note: You must unexport a LUN prior to deleting it. You are not allowed to delete an exported

LUN.

# no lun export container name

Showing a List of Thin Provision LUNs You can view information about thin provision and thick LUNs using the show luns command. You can filter LUNs by container, name, and session parameters. To view a list of containers, enter show luns container ?. The name parameter enables you to view only those LUNs which are prefixed by a particular name.

262


535-0008-00 Rev 10

Snapshots

# show luns container name sessions The following example shows the results for thin and thick LUNs. # show luns Container: MyContainer LUN Size Type RW Bksz Status Ports --------------------------------------------------------------------Mylun1 10G thin rw 512 HA A,B Mylun2 5G thick rw 512 HA A,B

Deleting a Thin Provision LUN The delete LUN command is the same for thin and thick (regular) LUNs.


LUN.

# no lun create container name

Snapshots A snapshot is a space-efficient, point-in-time copy of a LUN or other storage volume. By default, snapshots are read-only. However, you can set the permissions on snapshots to read-write. Snapshots help high-availability systems avoid down-time, because the time and I/O needed to create a snapshot does not increase with the size of the data set. A direct backup, however, is proportional to the size of the data set.

Note: An application has to be quiesced (on the client side) before you can take an application consistent snapshot. The snapshots discussed in this section refer to crash consistent snapshots, taken on the array.


• • • • • • • •

Creating Snapshots on page 264 Deleting Snapshots on page 264 Viewing Snapshot Information on page 265 Rolling Back the Contents of a Snapshot on page 267 Renaming Snapshots on page 268 Modifying the Access to Snapshots on page 269 Exporting Snapshots on page 269 Protecting Snapshots from Automatic Deletion on page 270

535-0008-00 Rev 10


263


Creating Snapshots Snapshot creation is instantaneous. When a snapshot is initially created it consumes no space. As writes occur to the LUN, the snapshot grows in size. The following rules apply for snapshots:

• • • •

A LUN can have a maximum of 1,024 snapshots

•

Deleting a single snapshot from a snap group is not allowed

A container can have a maximum of 10,000 snapshots across all the LUNs in the container Taking a snapshot of a snapshot is not allowed Snapshots are taken one at a time (serial); creating multiple simultaneous snapshots is not allowed at this time

Caution: Thin provisioned LUNs and LUNs with snapshots should always use ALUA, to ensure

optimum performance. For more information, see Optimizing Performance with ALUA on page 257.

By default, snapshots are created read-only. However, you can choose to create read-write snapshots, or change a snapshot to a read-write after it’s created. A snapshot name must start with alphanumeric characters, and may also include dashes and underscores. The [description] option allows you to add a description to the snapshot. You can allow write privileges to the snapshot with the [readwrite] option. The [protect] option prevents the snapshot from being deleted during space reclamation. # snapshot create container lun name [description][readwrite][protect]

Deleting Snapshots You must have Admin user privileges to delete snapshots. When a snapshot is deleted, the amount of space allocated to the snapshot is returned to free space. Review the following rules apply when deleting snapshots:

•

Snapshots that belong to a group cannot be deleted individually. However, you can delete all the snapshots in a group.

•

Snapshots that have been exported cannot be deleted. You must unexport the snapshot first, then it can be deleted.

•


# no snapshot create container lun name

264


535-0008-00 Rev 10

Snapshots

Viewing Snapshot Information You can display the same information for snapshots as you can for LUNs. For more information, see Viewing LUNs and Containers on page 192 and Optimizing Performance with ALUA on page 257. The show snapshots command has the following options: — — — — — — — — — —

alua: displays snapshots and their associated ALUAs container: displays the snapshots within a specific container count: displays snapshots lun: displays only the snapshots within a specific LUN naca: displays snapshots and their associated NACAs preferredport: displays snapshots and their associated ALUA preferred ports reserved: displays snapshots with SCSI reservations serial: displays the serial numbers for each snapshot sessions: displays snapshot sessions udid: displays snapshots and their associated UDIDs

Showing a List of Snapshots Use the following command to display a list of snapshots on the array. Along with the list of snapshots, the name, timestamp, read-only/read-write status, and protected information for each snapshot is displayed. # show snapshots

Container: 41238F00608 Snapshot LUN RW Protect Status Ports AllocSz Created --------------------------------------------------------------------------------------snapg1_lun1_20130416_133533 lun1 ro no HA A,B 0G 2013/04/16_13:35:33 snapg1_lun2_20130416_133533 lun2 ro no HA A,B 0G 2013/04/16_13:35:33 snapg1_lun3_20130416_133533 lun3 ro no HA A,B 0G 2013/04/16_13:35:33

Showing a List of Snapshots for a LUN Use the following command to display a list of snapshots for a specific LUN. # show snapshots container lun

Showing a List of Snapshot Sessions Use the following command to display a list of snapshot sessions: # show snapshots sessions

535-0008-00 Rev 10


265


Container: MyContainer Snapshot LUN RW Protect Status Ports AllocSz ---------------------------------------------------------------------s3-20-13 lun3 ro no HA A,B 50G Connected Sessions : vmem-dm1-mga : Port: hba-a2, Session: wwn.21:00:00:24:ff:26:68:08, LUN ID: 1 : Port: hba-a2, Session: wwn.21:00:00:24:ff:26:68:64, LUN ID: 1 : Port: hba-a2, Session: wwn.21:00:00:24:ff:26:68:09, LUN ID: 1 Connected Sessions : Port: hba-a2, : Port: hba-a2, : Port: hba-a2,

: vmem-dm1-mgb Session: wwn.21:00:00:24:ff:26:68:08, LUN ID: 1 Session: wwn.21:00:00:24:ff:26:68:64, LUN ID: 1 Session: wwn.21:00:00:24:ff:26:68:09, LUN ID: 1

Showing Snapshots with Associated ALUAs Use the following command to display a list of snapshots and their associated ALUAs. For information on ALUA and preferred ports, see Optimizing Performance with ALUA on page 257. # show snapshots alua Container: MyContainer snapshot ALUA ------------------------------vmem_20130319_175400 0 lun1_20130319_175700 1 lun2_20130319_180000 1 lun3_20130319_180301 0 lun4_20130319_180601 0 HR_1_20130319_181201 0 Eng1_20130319_181500 1

Showing Snapshots with Associated NACAs Use the following command to display a list of snapshots and their associated NACAs. For more information on NACA, see AIX Compatibility NACA Bit Support on page 179. # show snapshots naca

266


535-0008-00 Rev 10

Snapshots

Container: MyContainer snapshot NACA ------------------------------Admin_20130329_5500 0 lun5_20130329_5700 1 lun7_20130329_5800 1 lun3_20130329_5301 0 lun9_20130329_5601 1 HR_1_20130329_5201 0 Eng1_20130329_5900 1

Showing a List of Exported Snapshots Use the following command to display a list of exported snapshots. # show exports

Container: MyContainer LUN Snapshot Target Initiator LUNID Status --------------------------------------------------------------------Admin1_thick 20813 iscsi iqn.1994-05.com:01f775debf9 1 Active lun05_thin 20814 iscsi iqn.1994-05.com:01f775debf9 3 Active lun09_thin 20815 iscsi iqn.1994-05.com:01f775debf9 4 Active HR13_thick 20816 iscsi iqn.1994-05.com:01f775debf9 2 Active

Rolling Back the Contents of a Snapshot A rollback is the process of restoring the contents of a LUN from a snapshot. You can perform this task without unexporting an exported LUN.

Performing a Rollback Before you try performing a rollback, you should understand the requirements and the process. Requirements: a. Both ports of a LUN must be offline before issuing the “snapshot rollback” command. b. If the snapshot you want to restore the LUN from is read-write, the snapshot must be

unexported first. c. After the snapshot is unexported, the LUN you want to restore must be unexported as well.

535-0008-00 Rev 10


267


Rollback Process:

A rollback follows this process: a. When a rollback is initiated, the system creates a new snapshot of the LUN to preserve

the state of the LUN prior to the rollback. b. The LUN is changed to match the contents of the (restore) snapshot specified in the rollback. c. The LUN can be brought back online, as well as the (restore) snapshot used in the rollback process. Note: The system snapshot—taken at the beginning of the rollback process—can later be used

to undo the rollback, should the need arise.

Command Syntax Offline Ports

# no lun set container name port online Snapshot Rollback

# snapshot rollback set container lun name The following example shows a snapshot rollback command, preceded by both ports of a LUN being taken offline. violin-123 [violin: master] (config) # no lun set container 123 name lun1 port A online Data disruption will occur. Really offline LUN lun1 on port A? [no] yes Success violin-123 [violin: master] (config) # no lun set container 123 name lun1 port B online Data disruption will occur. Really offline LUN lun1 on port B? [no] yes Success violin-123 [violin: master] (config) # snapshot rollback container 123 lun lun1 name s1 WARNING! The contents of LUN lun1 will be overwritten with the contents of snapshot s1. Do you want to continue? [no] yes

Renaming Snapshots You can change the name of a snapshot at any time, provided it has not been exported. Changing the name of a snapshot does not impact the existing sessions for that snapshot, as long as the snapshot has not yet been exported.

Note: Changing the snapshot name after the snapshot has been exported is not recommended,

because traffic can be disrupted. The effect on the client depends on how it handles the change.

Specify the container ID, LUN name, old snapshot name, and new snapshot name.

268


535-0008-00 Rev 10

Snapshots

# snapshot rename container lun name to

Modifying the Access to Snapshots You can change snapshot access from read-only to read-write, or from read-write to read-only.

Note: Existing snapshot sessions are disconnected after making this change.

You are not allowed to modify the individual properties of a snapshot that belong to a snap group. For example, you are not allowed to change the read-only/read-write property or the protect property of a snapshot that belongs to a snap group. # snapshot set container lun name readwrite

Exporting Snapshots You can export snapshots in the same way you can export LUNs, using the following options: — lunid: assigns a LUN ID to the exported snapshot — to: exports a snapshot to an initiator or initiator group — using: exports a snapshot using a target port

Basic Syntax for Exporting a Snapshot # snapshot export container lun [snapshotname] [lunid] [to ]... [using ]

Assigning a LUN ID to an Exported Snapshot # snapshot export container lun [snapshotname] [lunid]

Exporting a Snapshot to an Initiator or Initiator Group # snapshot export container lun [snapshotname] [to ]

Exporting a Snapshot Using a Target Port # snapshot export container lun [snapshotname] [using ]

Unexporting a Snapshot # no snapshot export container lun [snapshotname]

535-0008-00 Rev 10


269


Protecting Snapshots from Automatic Deletion You can protect snapshots from being deleted by the system during space reclamation. A protected snapshot can never be deleted by the system, it must be manually deleted by an Admin user. The system can only delete unprotected snapshots to gain free space.

Note: If there are no unprotected snapshots and free space has gone below the threshold value,

an event is generated using syslog, SNMP trap, email and other supported notification methods.

Protecting a Snapshot # snapshot set container lun name protect

Unprotecting a Snapshot # no snapshot set container lun name protect

Snap Groups A snap group is an object that represents a set of snapshots taken across a collection of LUNs that belong to a LUN group. When a snapshot of a LUN group is taken, a snapshot is taken of each individual LUN in the LUN group (serially), then all the resulting snapshots are grouped together as a Snap Group. Snap groups are useful for creating an application level checkpoint, when an application is deployed across several LUNs. This section covers the following topics:

• • • • • • •

Creating a Snap Group on page 270 Deleting a Snap Group on page 271 Showing a List of Snap Groups on page 271 Rolling Back the Contents of a Snap Group on page 272 Modifying a Snap Group on page 272 Changing the Name of a Snap Group on page 273 Exporting a Snap Group on page 273

Creating a Snap Group You can create a snap group from an established LUN group. A snap group is a group of snapshots of the LUNs belonging to the LUN group. This operation creates a snapshot of each LUN in the snap group, then links the snapshots together into a group. # snapgroup create container lungroup name [readwrite] [protect] [description]

270


535-0008-00 Rev 10

Snap Groups

For information on how to modify a snap group after it’s created, see Modifying a Snap Group on page 272.

Deleting a Snap Group You can delete a snap group using its name. When you delete a snap group, all the snapshots contained within the group are deleted.

Note: Snapshots that belong to a group cannot be deleted individually. However, you can delete

the entire group of snapshots. # no snapgroup create container lungroup name

Showing a List of Snap Groups You can display a list of the snapshots in a snap group using any of the following options: — container: displays only the snap groups in a specific container. — LUN group name: displays only the snap groups that belong to a specific LUN group. — snap group name: displays only a specific snap group.

Viewing Snap Group Information In the following command, if you leave out the snapgroup-name, the resulting list shows all snap groups for the LUN group on the container. If you leave out the lungroup-name and snapgroupname, the resulting list displays all snap groups for all LUN groups on the container. If you use the show snapgroups command without any options, the resulting display shows the snap groups for all LUN groups, along with the associated snapshots. # show snapgroups [container] [lungroup] [snapgroup]

# show snapgroups container 41238F00608 lungroup lung1 snapgroup snapg0 Container: 41238F00608 Lungroup: lung1 Description: Just another snapgroup # of snapshots: 3 Snapshots ---------------------------------------------------------------------snapg1_lun1_20130416_133533 snapg1_lun2_20130416_133533 snapg1_lun3_20130416_133533 Total Allocated Size: 0G Snapshot Protected/Unprotected: 0G/0G

535-0008-00 Rev 10


271


Rolling Back the Contents of a Snap Group You can roll back (restore) the contents of all LUNs in a LUN group from a snap group. This operation does not require that the LUNs be unexported. However, all the LUNs and snapshots involved are taken offline when the contents of the LUN are changed. A LUN group may change after a snap group is created. In this case, the LUNs that were in the LUN group when the snap group was created are used for the rollback. If a LUN no longer exists, it is skipped. Before beginning a rollback, the system creates a new snap group to preserve the contents of the LUNs. If a rollback of a LUN fails, the operation is aborted.

Note: Each LUN must have port A and port B put offline before a rollback can be executed.

Rolling Back a Snap Group The following command specifies the LUN group to be restored by the specified snap group. # snapgroup rollback container lungroup name

Modifying a Snap Group This section demonstrates how to change the characteristics of a snap group.

Adding or Changing a Description Use the following command to add a description to a snap group, or change an existing description. Enter the new description text in place of the variable. The variable can contain multiple words if the words are enclosed within double quotes. # snapgroup set container lungroup name description

Protecting and Unprotecting a Snap Group Use the following command to protect a snap group from being automatically deleted when free space is needed. # snapgroup set container lungroup name protect Use the following command to allow the snap group to be automatically deleted when free space is needed. # no snapgroup set container lungroup name protect

272


535-0008-00 Rev 10

Snap Groups

Specifying Read/Write Permissions Use the following command to grant read/write access to a snap group. # snapgroup set container lungroup name readwrite Use the following command to set the permissions to read-only for a snap group with read/write access. # no snapgroup set container lungroup name readwrite

Changing the Name of a Snap Group You can change the name of a snap group at any time.

Renaming a Snap Group Use the following command to change the name of an existing snap group. You must specify the LUN group associated with the snap group. # snapgroup rename container lungroup name to

Exporting a Snap Group You can export a snap group and all the snapshots contained within the snap group, using the following options. — to: exports a snapshot to an initiator or initiator group — using: exports a snapshot using a target port

Exporting a Snap Group to an Initiator Use the following command to export a snap group to a specified initiator. # snapgroup export container lungroup name [snap_group_name] to

Exporting a Snap Group Using a Port Use the following command to export a snap group using a specified port or ports. # snapgroup export container lungroup name [snap_group_name] using

535-0008-00 Rev 10


273


Unexporting a Snap Group Use the following command to unexport a snap group. # no snapgroup export container lungroup name [snap_group_name]

Schedule Management You can automatically schedule the creation of snapshots and snap groups using schedule management commands. This feature allows you to define a schedule that creates snapshots or snap groups at specified time intervals. Every schedule supports a maximum number of snapshots and snap groups that can be kept at one time, through the max-keep option. When the max-keep option is implemented, if the maximum number of stored snapshots is reached, the oldest unexported and unprotected snapshot or snap group (created by the scheduler) is deleted when the next snapshot is created.

Note: The max-keep limit is set on a per schedule basis, not per LUN or LUN group.


• • • •

Creating a Schedule on page 274 Deleting a Schedule on page 275 Modifying Schedules on page 276 Showing a List of Schedules on page 279

Creating a Schedule Snapshots can be taken at regular intervals to facilitate space efficient backups. Schedules automate this process, offering a variety of options so you can tailor them to different needs. For information on how to create a schedule using the Web interface, see Creating Schedules on page 85. You can create more than one schedule for a given LUN or LUN group. However, each schedule applies to only one LUN or LUN group. A cluster can support up to 1024 schedules. When multiple schedules are set to be executed at the same time, they execute one after the other, not simultaneously. The sequential execution of schedules may cause a delay between the specified time of execution and the actual time the schedule is launched. Each snapshot created by the scheduler is created using the following naming convention: __ Note: If you change the name of a LUN, any existing schedules that include the LUN will not

reflect the name change. LUN names are not automatically changed in schedules, however, the snapshot schedule will continue un-impacted with the renamed LUNs.

274


535-0008-00 Rev 10

Schedule Management

By default, a schedule snapshot is created as read-only. However, when creating a schedule, you can change this option to read-write.

Customizing a Schedule Use the following command syntax to create a schedule: # schedule snapshot create container # write memory Use the following to customize a schedule: — description: Adds an identifying description to the schedule — disabled: Schedule will be disabled upon creation; default is enabled; it can later be modified to be enabled — end-date: Specifies the end date for the schedule — end-time: Specifies the end time for the schedule — lun: Creates a schedule for a LUN — lungroup: Creates a schedule for a LUN group — max-keep: Specifies the number of snapshots kept before deleting; does not apply for a one-time schedule — protect: Specifies the snapshots taken by this schedule are protected from automatic deletion — readwrite: Specifies that the snapshots taken by this schedule have read/write access — schedule-name: Specifies a name for the schedule — start-date: Specifies the start date for the schedule — start-time: Specifies the start time for the schedule — time-of-day: Specifies the time of day at which a non-periodic schedule triggers — type-daily: Specifies a daily schedule [time-of-day] — type-monthly: Specifies a monthly schedule [day-of-month] [interval] [day-of-month]: Specifies the day of the month on which a monthly schedule executes; use a negative number to count backwards from the end of the month [interval]: Specifies the number of months for the schedule to repeat; the default is 1 if this option is not specified — type-once: Specifies a one-time schedule [date] [date]: Specifies the execution date for a one-time schedule — type-periodic: Specifies a periodic schedule [periodicity] [periodicity]: Specifies the number of recurring intervals for the schedule — type-weekly: Specifies a weekly schedule [days-of-week] [days-of-week]: Specifies the days of the week on which a weekly schedule executes

Deleting a Schedule You can delete a schedule at any time. Deleting a schedule does not affect (or impact in any way) the snapshots or snap groups already created by the schedule. If a schedule is currently in the process of creating a snapshot or snap group, it does not impact the creation. When deleting a schedule, you have the option of deleting all unprotected and unexported snapshots and snap groups created by that schedule. When a LUN or LUN group is deleted, all the schedules associated with that LUN or LUN group are automatically deleted. 535-0008-00 Rev 10


275


For information on how to delete a schedule using the Web interface, see Modifying and Deleting Schedules on page 88.

Schedule Delete Command Specify the container name, the LUN or lungroup name and the schedule name to delete a schedule. # no schedule snapshot create container For example: # no schedule snapshot create container 41238F00608 lun lun1 schedule-name sched1 # write memory

Modifying Schedules Every schedule supports a maximum number of snapshots and snap groups that can be kept at one time, through the max-keep option. When the max-keep value is decreased, unexported and unprotected snapshots and snap groups are deleted as necessary, to meet the new max-keep requirement. The default value of max-keep is 20 snapshots. For information on how to modify schedules using the Web interface, see Modifying and Deleting Schedules on page 88. This section covers the following topics:

• • • • • • • • • • •

Modify Schedule Options and Examples on page 276 Basic Syntax for Modifying a Schedule on page 277 Changing a Schedule Name on page 277 Changing the Maximum Number of Kept Snapshots on page 277 Enabling a Schedule on page 277 Disabling a Schedule on page 277 Making a Schedule Read-Only on page 278 Making a Schedule Read-Write on page 278 Protecting Snapshots in a Schedule on page 278 Unprotecting Snapshots in a Schedule on page 278 Changing the Time of Day a Schedule Executes on page 278

Modify Schedule Options and Examples The command options, as described in Customizing a Schedule on page 275, also apply to modifying a schedule. The following options are unique to modifying a schedule: — enable: Activates a schedule — disable: Deactivates a schedule — protect: Protects the snapshots created by the schedule from automatic deletion by the system to create free space

276


535-0008-00 Rev 10

Schedule Management

— unprotect: Allows the snapshots created by the schedule to be automatically deleted by the system to create free space — readonly: Sets the permissions for the snapshots created by the schedule as read-only — readwrite: Sets the permissions for the snapshots created by the schedule as read-write Note: You can disable—temporarily pause—a schedule at any time. If a schedule is in progress,

the disable action does not interrupt or abort the schedule. The schedule will be put into a disabled state after the creation is complete.

The following sections demonstrate the basic syntax for modifying a schedule, and provides examples of common uses as well as the options that are unique to modifying a schedule.

Basic Syntax for Modifying a Schedule Use the following command with the necessary options to modify an existing schedule. For information on available options, see Customizing a Schedule on page 275. # schedule snapshot modify container lun|lungroup

Changing a Schedule Name The following example changes the name of a schedule associated with the specified LUN. # schedule snapshot modify container lun schedule-name new-schedule-name

Changing the Maximum Number of Kept Snapshots The following example changes the maximum number of snapshots that are kept. After this number is reached, the next oldest snapshot is deleted when a new snapshot is created. # schedule snapshot modify container lun schedule-name max-keep

Enabling a Schedule The following example enables a schedule. # schedule snapshot modify container lun schedule-name enable

Disabling a Schedule The following example disables a schedule. # schedule snapshot modify container lun schedule-name disable

535-0008-00 Rev 10


277


Note: You can disable—temporarily pause—a schedule at any time. If a schedule is in progress,


Making a Schedule Read-Only The following example sets the permissions for the snapshots created by the schedule as read-only. # schedule snapshot modify container lun schedule-name readonly

Making a Schedule Read-Write The following example sets the permissions for the snapshots created by the schedule as readwrite. # schedule snapshot modify container lun schedule-name readwrite

Protecting Snapshots in a Schedule The following example protects the snapshots created by the schedule so they are not automatically deleted by the system to create more free space. # schedule snapshot modify container lun schedule-name protect

Unprotecting Snapshots in a Schedule The following example allows the snapshots created by the schedule to be automatically deleted by the system to create more free space. # schedule snapshot modify container lun schedule-name unprotect

Changing the Time of Day a Schedule Executes The following example changes the time of day a schedule executes. # schedule snapshot modify container lun schedule-name time-of-day

278


535-0008-00 Rev 10

Free Space Management

Showing a List of Schedules You can display a list of schedule names using the “show schedule” command.

Viewing Schedules Use the following command to view a particular schedule. # show schedule If a snapshot is created by a schedule, it appears in the list of objects. In this case, the schedule name appears in the snapshot list.

Free Space Management You can monitor and manage the available space on containers for LUNs, snapshots, snap groups, and LUN groups. This section covers the following topics:

• • • •

Monitoring Free and Consumed Space on page 279 Rules for Reclaiming Free Space on page 283 Thresholds and the Reclamation Process on page 282 Setting Usage Thresholds on page 283

Monitoring Free and Consumed Space Monitoring free space (of a container) is recommended when using snapshots and snap groups. Monitoring free space is important when provisioned space goes beyond the physical or usable capacity of the array. Overprovisioning is the practice of provisioning space beyond the capacity of the array, since thin provision LUNs only use space as it is physically written. This section covers how to monitor free and consumed space in the following topics:

• • • • • •

Viewing a Container Space Summary on page 279 Viewing Container Space Utilization on page 280 Viewing LUN Space Utilization on page 280 Viewing Snapshot Space Utilization on page 281 Viewing Snapshot Space Utilization on page 281 Thresholds and the Reclamation Process on page 282

To monitor space through the Violin Web UI, see CHAPTER 2, “Manage, Monitor, and Administer with the Violin Web Interface” on page 31.

Viewing a Container Space Summary You can view a summary of the total space and free space on a container, along with the number of LUNs, and the provisioned space in Gigabytes. Ports, host, and status information is also displayed, using the following command. You must be in enable mode to run this command.

535-0008-00 Rev 10


279


# show containers Container LUNs Total Free Provisioned Ports HostA HostB Status ----------------------------------------------------------------41238F 8 20685G 20661G 40G A,B v-mga v-mgb HA

Viewing Container Space Utilization You can view a detailed breakdown of space utilization on a container with the show containers space_utilization command. The following terms apply:

•

provisioned size: amount of logical space used, or the space provisioned for use

•

allocated size: amount of physical space used

•

free space: amount of space available for use

You must enter enable mode to run this command. # show containers space-utilization name: MyContainer Category Allocated Provisioned ------------------------------------------------Thick LUNs 100G 100G Thin LUNs 640G 640G Snapshots 0 0 Reserved 30G Container Maximum space

770G

740G

20676G

20676G

Viewing LUN Space Utilization You can view space utilization data for individual LUNs as well as all LUNs on a container. This command displays detailed information about all the LUNs on a container. The following terms apply:

•


•

total (provisioned) size: amount of space asked for

You must enter enable mode to run these commands.

280


535-0008-00 Rev 10


# show luns Container: MyContainer LUN Size Type RW Bksz Status Ports Snapshots AllocSz TotalSz ---------------------------------------------------------------------lun5_thck 50G thick rw 512 HA A,B 324 50G 50G lun6_thck 75G thick rw 512 HA A,B 25 75G 75G lun7_thck 60G thick rw 512 HA A,B 202 60G 60G lun23_thn 35G thin rw 512 HA A,B 121 21G 23G lun24_thn 45G thin rw 512 HA A,B 221 12G 15G lun25_thn 50G thin rw 512 HA A,B 355 22G 27G The following command displays information about a particular LUN on a specified container. # show luns container name Container: MyContainer LUN Size Type RW Bksz Status Ports Snapshots AllocSz TotalSz ---------------------------------------------------------------------lun1 1G thin rw 512 HA A,B 1 0G 0G Total Allocated Size: 0G Lun Allocated/Provisioned Size: 0G/1G Lun Snapshot Protected/Unprotected/Shared Size: 0G/0G/0G

Viewing Snapshot Space Utilization You can view space utilization data for snapshots on a container. The following term applies:

•


You must enter enable mode to run this command. # show snapshots Container: 41238F00608 Snapshot LUN RW Protect Status Ports AllocSz Created --------------------------------------------------------------------------------------snapg1_lun1_20130416_133533 lun1 ro no HA A,B 0G 2013/04/16_13:35:33 snapg1_lun2_20130416_133533 lun2 ro no HA A,B 0G 2013/04/16_13:35:33 snapg1_lun3_20130416_133533 lun3 ro no HA A,B 0G 2013/04/16_13:35:33

535-0008-00 Rev 10


281


Thresholds and the Reclamation Process The following table lists the types of thresholds you can set to manage and reclaim free space, and the limits for each:

Threshold usedspace-threshold

Manages the physical space usage of the system

provision-threshold

Manages the “logical” usage of the system

Soft Limit

Hard Limit

Range: 0-100 Default: 0 => no used space alerts

Range: 0-100 Default: 0 => no space reclamation

When used space exceeds the soft limit, an SNMP alert is issued.

When used space exceeds the hard limit, snapshot space reclamation begins

Range: 0 - 999 Default: 0 => no provisioned space warnings or alerts

Range 0 - 999 Default: 100 => provisioned space equals usable space

When provisioned space exceeds the soft limit, LUN creation succeeds but a warning and SNMP alerts are issued.

When provisioned space exceeds the hard limit, further provisioning is stopped. The hard limit defines the “logical” size of the system.

Example

Soft limit = 50 Hard limit = 75 When 50% of the usable space in the array is used up, an SNMP alert is issued. When 75% of the usable space in the array is used up, space reclamation begins by deleting viable snapshots. Soft limit = 125 Hard limit = 150 Assuming the usable capacity of array is 20 TB, a hard limit of 150% allows the logical capacity of the array to be 30 TB. When creating a new LUN causes the total provisioned size to exceed 25 TB (125% of 20 TB), a warning is issued but the LUN creation succeeds. When creating a new LUN causes the total provisioned size to exceed 30 TB, LUN creation fails and an error is issued.

Table 8.1 Free Space Thresholds

Reclamation uses the following process: 1. Reclamation begins when the hard limit for the usedspace-threshold is exceeded. 2. The oldest to the newest unprotected, read-only, unexported snapshots are deleted. 3. When the amount of free space on the array goes below the trigger value or the max-delete

limit is reached (10 snapshots), reclamation halts. 4. If the max-delete limit is reached before the free space on the array drops below the trigger

value, reclamation resumes following the rules outlined in step 3. 282


535-0008-00 Rev 10


Rules for Reclaiming Free Space •

You should not specify all snapshots and snapgroups as protected, otherwise automatic space reclamation is not possible.

• •

You can never specify a LUN size larger than the total usable capacity on the container.

•

If the provision-threshold is exceeded, LUN creation is prohibited until LUNs are deleted or you increase the provisioning space threshold.

•

con The maximum number of snapshots that can be deleted at one time (max-delete default) is 10. The max-delete default cannot be modified.

•

Snapshots and snap groups are deleted from oldest to newest, until one of the following is achieved:

You can set the provision-threshold—the maximum provisioned space on the array—to a maximum of ten times (1000%) the physical space on the array.

— The free space on the array drops below the trigger value. — The maximum number (max-delete) of unprotected, read-only, unexported snapshots or snapgroups is deleted. — If the max-delete is reached and the available free space does not fall below the trigger, the identified snapshots and snap groups are deleted and the reclamation process is restarted.

•

You can specify the maximum number of snapshots or snap groups to keep (max-keep), as described in Modifying Schedules on page 276.

• •

A list of deleted snapshots is available in SNMP log files. If other policies are implemented by an external management software, you can set the reclamation thresholds to zero (0). This allows the external management software to be used instead of the Violin free space policies.

Setting Usage Thresholds You can specify space usage thresholds that activate alerts and space reclamation when the limit is reached. The threshold triggers are set at the container-level (array) and allow you to specify thresholds for free space and provisioned space. This section covers the following topics:

• • • •

Setting a Used Space Threshold Trigger on page 284 Setting a Provision Threshold Trigger on page 284 Deleting Threshold Triggers on page 284 Monitoring Available Free Space on page 284

For information on the rules that apply to space reclamation, see Rules for Reclaiming Free Space on page 283.

Note: Threshold values are in percentages (%). Threshold triggers are set at the container-level

(array).

535-0008-00 Rev 10


283


Setting a Used Space Threshold Trigger The following command specifies hard and soft thresholds (in percentages), to manage used space in the container. # container set name usedspace-threshold soft|hard

• •

soft: threshold issues an alert when the percentage is exceeded hard: threshold starts space reclamation when the percentage is exceeded

Setting a Provision Threshold Trigger The following command specifies hard and soft thresholds (in percentages), to manage provisioned space in the container. # container set name provision-threshold soft|hard

• •

soft: threshold issues an alert when the percentage is exceeded hard: threshold prevents further provisioning when the percentage is exceeded

Deleting Threshold Triggers The following command deletes a used space threshold set for a specific container. # no container set name usedspace-threshold [soft] [hard] The following command deletes a provisioned space threshold set for a specific container. # no container set name provision-threshold [soft] [hard]

• •

soft: threshold issues a warning when the percentage is exceeded hard: threshold starts reclamation when the percentage is exceeded

Monitoring Available Free Space The commands for monitoring the available free space on a container or array are covered in Monitoring Free and Consumed Space on page 279. For information on how to monitor free space in at-a-glance visual displays, see Space Management on page 102.

284


535-0008-00 Rev 10

REFERENCE

APPENDIX A, “CLI Reference” on page 287 APPENDIX B, “Alarm Reference” on page 369 APPENDIX C, “Violin Drivers” on page 387 APPENDIX D, “Violin Utilities” on page 403 APPENDIX E, “Performance Tuning and Testing” on page 423 APPENDIX F, “SNMP Traps” on page 433 APPENDIX G, “Optimizing Connectivity for Initiator Hosts” on page 465

CLI Reference

APPENDIX A

This appendix lists and describes the available commands in the command line interface (CLI) of the Violin 6000 Series Memory Array in the following sections.

• • • • • • • • • • • • • • • • • • • • •

535-0008-00 Rev 10

CLI Command Descriptions on page 288 CLI Option Commands on page 289 General Configuration Commands on page 292 Network Configuration Commands on page 314 Administrative Commands on page 316 Email Alert Consolidation Commands on page 317 Unified Logging Upload Commands on page 319 Address Resolution Protocol Commands on page 319 Array Configuration and Show Commands on page 320 VIMM Commands on page 324 iSCSI Commands on page 325 Security Commands on page 326 SCSI-3 ALUA Commands on page 330 Fibre Channel Commands on page 331 Initiator Group Commands on page 331 LUN Commands on page 333 LUN Group Commands on page 338 Data Management Commands on page 341 Diagnostic and Display Commands on page 359 Standard Show Commands on page 359 Standard CLI Commands on page 361


287

Appendix A - CLI Reference

CLI Command Descriptions The sections below describe CLI commands for the Violin 6000 Series Memory Array.

• • •

Key to Command Parameters on page 288 General Configuration Commands on page 292 Network Configuration Commands on page 314

Key to Command Parameters This section is a key to the meaning and format of parameter values and other attributes of the CLI commands. Parameter values are shown in angle brackets, and listed alphabetically below.

288

A string specifying the name of a cluster.

A domain name, such as “vmem.com”.

A hostname, such as “hexagon.vmem.com”.

An interface name, such as “eth0”, “eth1”, “lo” (loopback), and so on.

An IPv4 address, such as “192.168.0.1”.

A syslog logging severity level. Possible values, from least to most severe, are: “debug”, “info”, “notice”, “warning”, “error”, “crit”, “alert”, “emerg”.

A MAC address. The segments may be 8 bits or 16 bits at a time, and may be delimited by “:” or “.”. So you could say “11:22:33:44:55:66”, “1122:3344:5566”, “11.22.33.44.55.66”, or “1122.3344.5566”.

A netmask (such as “255.255.255.0”) or mask length prefixed with a slash (such as “/24”). These two express the same information in different formats.

An IPv4 network prefix specifying a network. This is used in conjunction with a netmask to determine which bits are significant. For example, “192.168.0.0”.

ID of a node (Memory Gateway, Violin 6000 Series Memory Array) belonging to a cluster. This is a numerical value greater than zero.

A TCP or UDP port number.

An extended regular expression as defined by the “grep” man page. (The value you provide here is passed on to “grep -E”.)

A TCP port number in the full allowable range [0...65535].

Either a normal URL, using any protocol that wget supports, including http, https


535-0008-00 Rev 10

CLI Option Commands

If you omit the “:password” part, you may be prompted for the password in a follow up prompt, where you can type it securely (without the characters being echoed). This prompt will occur only if the “cli default prompt empty-password” setting is true; otherwise, the CLI assumes you do not want any password. If you include the “:S” character, this is taken as an explicit declaration that the password is empty, and you will not be prompted in any case. INTERACTIVE

A command that is only available for interactive usage from the CLI shell. Specifically, this excludes using the command from jobs, because the Scheduler runs commands in a batch mode with no direct user interactivity. In most cases, it is obvious why each of these commands has the INTERACTIVE restriction.

CLI Option Commands The following groups of commands relate to the CLI itself:

•

cli session ... commands change a setting only for the current CLI session. They do not affect any other sessions, and can be performed by any user at any time. All of these commands are INTERACTIVE (not available from the Scheduler).

•

cli default ... commands change the defaults for the specified setting for all future CLI sessions of all users. They also change the setting for the current session from which they were executed, but not for any other currently active sessions. Since they change configuration, the user must be in Configure mode to run them, and hence they can only be run by admin.

•

Other cli ... commands which take one-time actions, rather than change a setting, and thus do not fall under the session or default umbrellas. For example, cli clear-history.

•

terminal ... commands are clones of a subset of the cli session ... commands, and are only present for Cisco compatibility. All of these commands are INTERACTIVE (not available from the Scheduler).

Note that some settings, such as the terminal length and width, are inherently session-specific, and there are no corresponding commands to set the defaults in configuration.

Control User Inactivity cli default auto-logout cli session auto-logout no cli default auto-logout no cli session auto-logout Control the length of user inactivity required before the CLI will automatically log a user out. The no ... variants of this command disable the automatic logout feature.

535-0008-00 Rev 10


289


Control Paging of CLI Output [no] cli default paging enable [no] cli session paging enable Enable or disable paging of CLI output. If paging is enabled, all command output, as well as all help text printed when the ? key is pressed, will be displayed one screen at a time, using the same pager as the 'show log' command uses. If the text to be displayed fits on a single screen, it will be printed normally and the pager will not be used. Note that the abbreviated list of commands printed when is hit twice is not paged, even in the unlikely event that it does not fit on the screen. Note that if the CLI does not have a terminal (for example, it is being driven by a script), paging is disabled automatically regardless of the default setting, and cannot be re-enabled for this session. However, even in this case, the default setting can still be changed.

Control Showing Hidden Commands [no] cli default show config-hidden enable Enable or disable showing hidden commands in the output of the various forms of show config .... By default, hidden commands are displayed to represent the running state of a system.

Confirm Loss of Unsaved Changes [no] cli default confirm lose-unsaved Enable or disable confirmations of cases where you might accidentally lose unsaved changes. Currently, this is just for the reload [halt] command; other cases where you might lose configuration are just some of the configuration ... commands, which have no confirmations in any case, since they are explicitly dealing with configuration.

Confirm Reboot or Halt Reload [no] cli default confirm reload Enable or disable confirmations of rebooting or halting the system using the 'reload' command. This confirmation is in addition to any separate confirmations that may be displayed for unsaved changes. cli session terminal width cli session terminal length terminal width terminal length Override the auto-detected size of the terminal. This is useful mostly when the size could not be auto-detected and the CLI is using the default 80x24. 290


535-0008-00 Rev 10

CLI Option Commands

These settings are persistent only for the current CLI session. They are also lost if the terminal is resized and the CLI is able to auto-detect its new size. Note that the commands without the cli session prefix are identical to those with the prefix, and are present only for Cisco compatibility.

Set Terminal Type cli session terminal type no cli session terminal type terminal type no terminal type Set the type of the terminal. The 'no' variants clear the terminal setting, which will cause the session to be treated as a “dumb” terminal. The commands without the cli session prefix are identical to those with the prefix, and are present only for Cisco compatibility.

Clear Command History cli clear-history Clears the command history of the current user.

Display CLI Settings show cli Display CLI settings: the inactivity timeout, whether or not paging is enabled, the terminal size and type. For settings which have configured defaults, both those and the current session settings are displayed.

Display Terminal Details show terminal Display current terminal width and length, whether auto-detected or overridden, as well as the current terminal type. This is a subset of the information displayed by show cli, and is only present for Cisco compatibility.

535-0008-00 Rev 10


291


General Configuration Commands This section describes CLI commands for these tasks:

• • • • • • • • • • • • • •

Verify Hardware Version and Capabilities on page 292 View Installed Licenses on page 294 Event Logging Configuration and Viewing on page 294 User Accounts and Local Authentication on page 299 NTP, Clock and Time Zones on page 300 Event Notification on page 302 Diagnostic Tools on page 303 Configuration Commands on page 305 Enable and Config Modes on page 310 Web Proxy Settings on page 311 Xinetd Server Settings on page 311 tcpdump Capture Files on page 311 NTB Commands (vMOS Data Management) on page 312 Cluster Configuration and Show Commands on page 313

Verify Hardware Version and Capabilities Some software features are optimized for a particular Violin Memory platform. This section demonstrates how to verify the hardware version of a system and its inherent capabilities.

Show Hardware Version show version

292


535-0008-00 Rev 10

General Configuration Commands

Displays the product information for the system you are logged in to, as shown in the following example. vmem-mga [cluster: master] # show version Product name: Product release: Build ID: Build date: Target arch: Built by: Array version: Array Build ID: Cluster ID: Licensed for:

vMOS V6.0.0 #27 2013-04-24 20:13:18 x86_64 vmem-builds.int A6.0.0 #12 99999-9999-4583 Violin Memory Gateway [ VMOS6 ]

Uptime:

7d 19h 9m 54.688s

Product model: Host ID: Serial number:

tt_shb 9d959bb327b4 47208T00102

Show Hardware Capabilities show capabilities Displays the capabilities (of the hardware) of the system you are logged in to, without discerning if the feature is enabled or activated. The following example shows that the Internal Gateway of the Violin 6000 Series flash Memory Array is capable of Encryption. However, to use the Encryption feature a special license must be installed. For information on how to verify if a license is installed, see View Installed Licenses on page 294. Contact Violin Memory Customer Support to purchase a license, as described in Contacting Violin Memory on page 5. *vmem-mg-b [vmem: master] (config) # show capabilities Internal Gateway: yes Encryption Support: yes Serial Redirect: no USB Boot : no *vmem-mg-b [vmem: master] (config) #

535-0008-00 Rev 10


293


View Installed Licenses show license Displays licenses installed on the system, along with associated license key. If an installed license shows No as its Active output, an error may have been made when inputting the license key. Double-check the valid license key against the key that is shown in the show license output.

Event Logging Configuration and Viewing This section provides the syntax for event log commands, along with an explanation of the functionality of each command.

Set Local Log Severity logging local no logging local Set the minimum severity of log messages to be saved in log files on local persistent storage, disable local logging altogether. The logging local none and no logging local commands have the same effect. [no] logging Send syslog messages to a remote syslog server, or stop sending messages to the specified server. Note that hostnames are not allowed here.

Set Log Severity for All Syslog Servers logging trap Set the minimum severity of log messages sent to all syslog servers. Note that the Cisco command set does not provide for setting different severity levels on a per-server basis. Set, clear, enable, or disable class-specific log level overrides for all syslog servers. Semantics are the same as for the logging local ... variants above. [no] array serial-logging Enable or disable serial logging of vRAID Controller Modules and Memory Gateways.

Configure a Custom Syslog Port vsyslog port logging The vsyslog port command works in conjunction with logging command to set up the forwarding of syslog to remote host. The default syslog forwarding port number is 514. To forward the syslog to a different port, issue the vsyslog port command before the logging command, as shown in the following example. myvmem-n1 [myvmem-n1: master] (config) # vsyslog port 5555 myvmem-n1 [myvmem-n1: master] (config) # logging 10.1.4.169 myvmem-n1 [myvmem-n1: master] (config) # write mem

Specify Log Overrides logging local override class priority no logging local override class Set or remove a per-class override on the logging level. All classes that do not have an override set will use the global logging level set with logging local ; any that do have an override will do as the override specifies. If none is specified for the log level, nothing from this class will be logged. Class is a user-friendly synonym for facility in syslog parlance. It allows log messages to be divided up according to their origin. The default Samara classes are mgmt-core (for mgmtd alone), mgmt-back (for other back end components), and mgmt-front (for front end components, utilities, and tests).

Enable or Disable Overrides [no] logging local override Enable or disable all class-specific overrides to the local log level. Overrides are enabled by default. The no variant which disables them will leave them in configuration, but disable them such that the logging level for all classes is determined solely by the global setting.

Send Logs to a Remote Server [no] logging Send syslog messages to a remote syslog server, or stop sending messages to the specified server. Note that hostnames are not allowed here.

Set Log Severity for a Specific Syslog Server logging trap Set the minimum severity of log messages sent to the indicated syslog server. This is not Cisco compatible (as noted above) but does match functionality provided in the reference UI.

Specify Log Overrides for a Specific Syslog Server logging trap override class priority no logging trap override class [no] logging trap override Set, clear, enable, or disable class-specific log level overrides for the specified syslog server. Semantics are the same as for the logging local ... variants above.

535-0008-00 Rev 10


295


Disable Logging for All Syslog Servers no logging trap Disable logging to syslog servers altogether. This simply sets the severity level to none; the list of servers is not erased. Does not affect console or local logging, despite the generic sound of the term trap.

Allow Logs from Another Host [no] logging receive Allow this system to receive log messages from another host. This is disabled by default. If enabled, only log messages matching or exceeding the minimum severity specified with the logging local command are logged, regardless of what is sent from the remote host.

Specify Log Format logging format standard logging format welf no logging format welf Sets the format in which log messages should be set. The choices are standard and welf. The default is standard, and the no variant resets the format to this.

Specify the Log Firewall logging format welf fw-name no logging format welf fw-name Specifies the firewall name that should be associated with each message logged in WELF format. If no firewall name is set, the hostname is used by default. Note that neither of these commands enables WELF logging if it is not already enabled.

Set CLI Command Severity logging level cli commands Set the severity level at which CLI commands that the user executes are logged. The default is info.

Specify When Logs are Rotated logging files rotation criteria frequency {daily, weekly, monthly} logging files rotation criteria size logging files rotation criteria size-pct Configure what criteria will be used to decide when log files on local persistent storage should be automatically rotated. There are two mutually exclusive options: rotate based on time, or based on when the active log file reaches a certain size threshold. If the latter is chosen, the size of the file is checked hourly, so if it passes the threshold in the middle of the hour it will not be rotated until the end of the hour.

296


535-0008-00 Rev 10


Specify Number of Old Log Files Saved logging files rotation max-num Configure how many old log files will be kept. If the number of log files ever exceeds this number (either at rotation time, or when this setting is lowered), the system will delete as many as necessary to bring it down to this number, starting with the oldest.

Force Log File Rotation logging files rotation force Force an immediate rotation of the log files. This does not affect the schedule of auto-rotation if it was done based on time: the next automatic rotation will still occur at the same time it was previously scheduled for. Naturally, if the auto-rotation was based on size, this will delay it somewhat as it reduces the size of the active log file to zero.

Force Deletion of Oldest Log Files logging files delete oldest [] Force the immediate deletion of the specified number of the oldest log files.

Upload Logs to a Remote Host logging files upload {current, } Upload a log file to a remote host. The word current specifies the current log file. To specify an archived log file, give its number instead, as displayed by show log files. The current log file will have the name messages if you don’t specify a new name for it in the upload URL. The archived log files will have the name messages..gz if you don’t specify a new name in the URL, and will be compressed with gzip regardless.

Note: Files are uploaded from all gateways simultaneously. Therefore, care should be taken to allow that many simultaneous connections on the server accepting the files (for example, increase the maximum open ftp connections parameter).

Enable Automatic Log Upload to Remote Host logging files upload-auto Enable automatic upload of log files to a remote host. The control parameters are used to control interval and information gathered. Control parameters are: Enable automatic log gathering.

enable immediate {current, }

535-0008-00 Rev 10

One-shot upload of logs to the configured site.


297


include-dump interval max-size remote-dir remote-site user

Enable inclusion of sysdump information during log gathering. Set how often to gather logs, in hours. Set the maximum uploadable file size, in MB. Set the remote directory for uploading logs. Set the remote site for uploading logs, in URL format. Change the user/password used with the remote server.

Note: Files are uploaded from all gateways simultaneously. Therefore, care should be taken to allow that many simultaneous connections on the server accepting the files (for example, increase the maximum open ftp connections parameter).

Display All Event Logs show logging Display all event logging configuration settings.

Display a Local Log File show log [files ] [matching ] View a local log file using the “less” pager.

•

If files is specified, view an archived log file, where the number is from 1 up to the number of archived log files.

•

If matching is specified, the file is piped through the grep utility to only include lines matching the provided regular expression.

Display the Last Lines of a Log Files show log continuous [matching ] Display the last few lines of the current log file, and then continue to display new lines as they come in, until the user presses CTRL+C. This is done using the tail utility. If matching is specified, only log lines matching the provided regular expression are printed.

298


535-0008-00 Rev 10


View a List of Local Log Files show log files View a list of local log files.

User Accounts and Local Authentication This section provides the syntax for commands used for creating and managing user accounts and local authentication. An explanation of the functionality of each command is also provided.

Note: There are two defined system accounts: one with full privileges (admin), and one with privileges for reading all data and performing all actions, but not for changing any configuration (monitor).

Create or Remove a User Account [no] username Create or remove a user account. New users are created initially with admin privileges and disabled. To enable a user account, just set a password on it (or use the nopassword command to enable it with no password required for login). Removing a user account does not terminate any current sessions that user has open; it just prevents new sessions from being established.

Change User Account Capabilities username capability no username capability Change the capabilities for this user account. Creates the account if it doesn’t exist. The system provides three predefined capabilities: admin, monitor, and unpriv. It is possible to statically expand this set. The no modifier reverts the specified user to the default capability, which is admin privileges.

Specify User Login Credentials User login credentials include a username and password.

Note: Two types of password encryption are supported, SHA-1 and MD5. Secure Hash Algorithms (SHA) are cryptographic hash functions structured differently and are distinguished by the appended number, such as -1. Message-Digest Algorithm is a cryptographic hash function that produces a 128 bit (16 byte) hash value.

username password username password 0 username password 7 535-0008-00 Rev 10


299


username nopassword [no] username disable [no] username disable password Control what it takes for the specified user to log in. The first three commands set a password on the account.

• •

The 0 flavor allows the password to be specified in cleartext. The 7 flavor allows the password to be provided in the same encrypted form in which it would be stored in the password file. This is useful for the show configuration command, since the cleartext password cannot be recovered after it is set. These two commands are thus named for Cisco compatibility.

The nopassword command means that no password is required to log in. The disable command configures the account so that no one can log into that account. The disable password command leaves the account as a whole the same, but forbids login with a password. It is assumed that ssh key access will be used instead. To re-enable the account, the user must un-disable it, and put a password on it. The no username disable command prints a message to this effect; it only exists to avoid stumping users with an apparently irreversible command. Disabling a user account does not terminate any current sessions that user has open; it just prevents new sessions from being established.

Display Current User Name show users Display the username of the currently logged-in user, and the capabilities that user has.

Display All User Accounts show usernames Display a list of all user accounts and the capabilities of each.

NTP, Clock and Time Zones clock set [] Set the system clock. The time must be specified. The date is optional; if not specified, the date will be left the same. clock timezone no clock timezone Set the system time zone. The time zone may be specified in one of three ways: A nearby city whose time zone rules to follow. The system has a large list of cities which can be displayed by the help and completion system. They are organized hierarchically because there are

300


535-0008-00 Rev 10


too many of them to display in a flat list. A given city may be required to be specified in two, three, or four words, depending on the city. The possible forms this could take include: An offset from GMT. This will be in the form GMT-offset GMT, GMT-offset GMT+, or GMT-offset GMT-. UTC. This is almost identical to GMT. The default is “GMT-offset GMT”. show clock Display the current system time, date, and timezone. ntpdate Set the system clock using the specified NTP server. This is a one-time operation and does not cause the clock to be kept in sync on an ongoing basis. It will generate an error if NTP is enabled, as the socket it requires will already be in use. ntp peer [version ] no ntp peer Add or remove an NTP peer. Allowable version numbers are 3 and 4. If no version number is specified when adding a peer, the default is 4. ntp server [version ] no ntp server Add or remove an NTP server. Allowable version numbers are 3 and 4. If no version number is specified when adding a server, the default is 4. [no] ntp peer disable [no] ntp server disable Disable or re-enable an NTP server or peer. Servers and peers start enabled; disabling is just a way of making them temporarily inactive without losing their configuration. [no] ntp enable [no] ntp disable Enable or disable NTP overall. Note that the latter is the Cisco command; the former is just a pair of aliases added to increase usability, as otherwise it may be hard for a user to figure out how to enable NTP if they are not aware of no commands and only see a way to disable it.

535-0008-00 Rev 10


301


show ntp Display current NTP settings.

Event Notification [no] email enable Enable or disable the sending of e-mail when certain events occur. These events are the same as those for which SNMP traps can be sent (see SNMP Configuration on page 365). email mailhub no email mailhub Set or clear the mail relay to use to send notification emails. email mailhub-port no email mailhub-port Set the mail relay port to use to send notification emails. The no variant resets the port to its default, which is 25. email domain no email domain Set the domain name from which the emails will appear to come from. This is used in conjunction with the system hostname to form the full name of the host from which the email appears to come. The rules are as follows:

•

If an email domain is specified using this command, it is always used. If the hostname has any dots in it, everything to the right of the first dot is stripped and the email domain is appended.

• •

Otherwise, if the hostname has dots in it, it is used as is. Otherwise, the currently-active system domain name is used. This can come either from the resolver configuration, or from state dynamically instantiated by DHCP.

[no] email notify event Enable or disable sending email notifications for the specified event type. The can be any of the following: process-crash, process-exit, liveness-failure, cpuutil-high, cpu-util-ok, paging-high, paging-ok, disk-space-low, disk-space-ok, memusage-high, memusage-ok, netusage-high, netusage-ok, disk-io-high, disk-io-ok, unexpected-cluster-join, unexpected-cluster-leave, unexpected-cluster-size, unexpected-shutdown, interface-up, interfacedown, user-login, user-logout, cache-faults, hwmon-file-sys-error, hwmon-mce-error, mediadevice-health-warn, media-device-state-change, media-device-lifetime-warn, media-deviceunknown-type, media-system-swap-state, excessive-shutdowns, license-state-change, conntrackentries-high, conntrack-entries-ok, vimm-temperature-high, vimm-temperature-ok, chassistemperature-high, chassis-temperature-ok, lid-ajar-time-high, lid-ajar-time-ok, array-pcie-link-up, array-pcie-link-down, array-data-plane-ready, array-raid-rebuild, vimm-state-change, array-psustate, array-led-change, fc-port-state-change, ib-port-state-change, array-fan-change, vmg-fanchange, vmg-psu-change, vmg-boot-raid-change, lid-ajar, kernel-crash, vmg-ha-split-brain, ipmidcoretemp, used-threshold-alert, used-threshold-limit, prov-threshold-alert, prov-threshold-limit.

302


535-0008-00 Rev 10


Note that this does not affect autosupport emails. Autosupport can be disabled overall, but if it is enabled, all autosupport events (currently only PM process failures) are sent as emails. [no] email notify recipient Add or remove an email address from the list of addresses to send email notifications of events. [no] email notify recipient class info [no] email notify recipient class failure Enable or disable the sending of informational or failure events to the specified recipient. Each event type is classified as either “informational” or “failure”. The specified recipient will receive the intersection of the set of events specified by this command, and the set of events specified overall with the [no] email notify event command. [no] email notify recipient detail Specify whether the emails this recipient is sent should be detailed or summarized. Each email potentially has both a detailed and summarized form, where the detailed form has a superset of the information. (In practice, only PM process failure emails currently have a detailed form; for everything else, the two are the same.) [no] email autosupport enable Enable or disable the sending of email to vendor autosupport when certain failures occur. email autosupport mailhub no email autosupport mailhub Set the mail relay to use to send autosupport emails. email send-test Send a test email to all of the configured notification email recipients. This is useful to make sure the configuration works without having to wait for an event to occur. show email Display notification settings. This does not include SNMP traps, which are under the snmp-server command tree.

Diagnostic Tools This section provides the syntax for a variety of diagnostic tool commands, along with an explanation of the functionality.

ping and traceroute ping [] traceroute [] Network diagnostic tools ping and traceroute. Invokes standard binary, passing command line parameters straight through.

535-0008-00 Rev 10


303


tcpdump tcpdump [] Network diagnostic tool tcpdump. Invokes standard binary, passing command line parameters straight through. Runs in foreground, printing packets as they arrive, until user presses CTRL+C.

slogin slogin [] Invokes the SSH client. The user is returned to the CLI when SSH finishes.

telnet telnet [] Invokes the telnet client. The user is returned to the CLI when telnet finishes.

show version show version [concise] Display version information for the currently running system image. The basic command shows each field with a description, one per line; the concise variant fits it all onto one line, without labels, in a form suitable for pasting in a bug report.

show files system show files system Display information regarding bytes and inodes usage of the file systems on the system.

show memory show memory Display information about system memory utilization.

reload reload [halt] [noconfirm] Reboot the system if reload, shut down the system if reload halt. If there are unsaved changes to the configuration, the user may be prompted as to whether they want to save these changes (that is, do a write memory) first before rebooting. The prompt will be suppressed if confirmation of losing unsaved changes is disabled (with the no cli default confirm lose-unsaved command). The user may also be prompted to confirm the reload regardless of whether there are unsaved changes or not. This prompt is contingent on a separate configuration setting, controlled with the 304


535-0008-00 Rev 10


[no] cli default confirm reload command. If both prompts are enabled, and the configuration was unsaved, the user will be prompted twice. The noconfirm parameter suppresses both of these confirmations if it is specified.

reload force reload force If the system is busy performing another operation requiring the management subsystem (which is almost any management operation), the regular reload [halt]command will block until it is finished. If rebooting the system is urgent, the reload force command will do it immediately. This reboots the system, and there is no halt variant. There is also never any confirmation, whether or not there are any unsaved changes to the configuration.

debug generate dump debug generate dump Generate a debugging dump. The dump can then be manipulated using the file debug-dump ... family of commands. debug metasave Enable metasave during sysdump.

Configuration Commands write memory Save the running configuration to the active configuration file. write memory local Same as “configuration write local”, provided for Cisco compatibility. write terminal Same as “show running-config”, provided for Cisco compatibility. configuration jump-start Run the initial-configuration wizard. The wizard is automatically invoked whenever the CLI is launched when the active configuration file is fresh (that is, not modified from its initial contents). This command invokes the wizard on demand. show configuration [full] Display a list of CLI commands that will bring the state of the a fresh system up to match the current persistent state of this system. A short header is included, containing the name and version number of the configuration, in a comment.

535-0008-00 Rev 10


305


Commands which are not required because they would set something to its default are not included – thus running this command on a fresh configuration will produce no output, aside from the header. Note that this does not include changes that have been made but not yet written to persistent storage. The “full” option will also include commands that set things to their defaults, but excludes those of them that are instances of hidden commands. show configuration running Same as “show configuration” except that it applies to the currently running configuration, rather than the current persisted configuration. “Show running-config” is for Cisco compatibility. show configuration files [filename] If no filename is specified, display a list of configuration files in persistent storage. If a filename is specified, display the commands to recreate the configuration in that file. In the latter case, only non-default commands are shown, as for the normal show configuration command; the full, all, and subtree options are not available.

Configuration File Management This section provides the syntax for configuration file management commands, along with an explanation of the functionality of each command. The system can store one or more configuration files on persistent storage. At any given time, one of the configuration files is designated as active. That is the file that configuration will be loaded from on boot, and which configuration will be saved to when a save is requested. Configuration changes are immediately applied to the running configuration (with the exception of configuration batch commands), but are not made persistent until they are explicitly saved using the configuration write command.

Note: These configuration commands only work when you are in Configure mode (which

you enter by using the conf t command—see Enable and Config Modes on page 310).

configuration new configuration new [factory [keep-basic]] Create a new configuration file under the specified name. If no additional parameters are specified, active licenses from the current running configuration are copied over into it. If the factory parameter is specified, nothing is copied over; it has the factory defaults. If the keep-basic parameter is specified, local network configuration is copied over so you won't lose network connectivity when you switch to it.

configuration revert configuration revert saved

306


535-0008-00 Rev 10


Revert the running configuration to the latest saved version of the active configuration file. configuration revert factory [keep-basic | keep-connect] Revert both running and saved configurations to factory defaults. If keep-basic is specified, preserve local network configuration. If keep-connect is specified, preserve active licenses. These two options are mutually exclusive.

configuration merge configuration merge Merges the shared configuration from one configuration file into the running configuration. No configuration files are modified during this process. The must be a non-active configuration file.

configuration switch-to configuration switch-to Load configuration from the specified file and change that to be the active configuration file. The current running configuration is lost, and not automatically saved to the previous active configuration file.

configuration write configuration write configuration write to Write the running configuration to persistent storage. The former command writes it to the currently active file. The latter command writes it to the specified file, and changes the active file to that one. configuration write local On a system with clustering, only save the configuration on the local box instead of attempting to save on all cluster members. configuration write to usb [] Save a local configuration file to a USB.

configuration delete, move, copy configuration delete configuration move configuration copy Delete, move (rename), or copy a configuration file. These do not affect the current running configuration. The active configuration file may not be deleted or renamed, nor may it be the target of a move or copy. It may be the source of a copy, in which case the original remains active.

535-0008-00 Rev 10


307


configuration fetch configuration fetch [] Download a configuration file from a remote host. configuration fetch usb [] Download a configuration file from a USB.

configuration upload configuration upload configuration upload active Download or upload a configuration file. A file may not be downloaded over the active configuration file. If no name is specified for a configuration fetch, it is given the same name as it had on the server. If active is specified for a configuration upload, the currently-active configuration file is uploaded. No configuration file may have the name active. See Key to Command Parameters on page 288 for a description of valid URLs.

Note: If downloading configuration files from another system, they can be found in the /config/

db directory. So an example command line to fetch the initial configuration database would be: configuration fetch scp://admin:password@hostname/config/db/initial write memory

Same as configuration write, provided for Cisco compatibility.

write memory local write memory local Same as configuration write local, provided for Cisco compatibility.

write terminal write terminal Same as show running-config, provided for Cisco compatibility.

configuration jump-start configuration jump-start Run the initial-configuration wizard. The wizard is automatically invoked whenever the CLI is launched when the active configuration file is fresh (that is, not modified from its initial contents). This command invokes the wizard on demand.

308


535-0008-00 Rev 10


show configuration show configuration [full] Display a list of CLI commands that will bring the state of the a fresh system up to match the current persistent state of this system. A short header is included, containing the name and version number of the configuration, in a comment. Commands which are not required because they would set something to its default are not included—thus running this command on a fresh configuration will produce no output, aside from the header. Note that this does not include changes that have been made but not yet written to persistent storage. The full option will also include commands that set things to their defaults, but excludes those of them that are instances of hidden commands. show configuration running Same as show configuration except that it applies to the currently running configuration, rather than the current persisted configuration. The command show running-config is for Cisco compatibility. show configuration files Displays a list of configuration files in persistent storage. show configuration files [filename] If a filename is specified, the commands to recreate the configuration in that file display. In the latter case, only non-default commands are shown, as for the normal show configuration command; the full, all, and subtree options are not available.

Image Management This section provides the syntax for image management commands, along with an explanation of the functionality of each command.

Specify Boot Location boot system location boot system next Specify which location the system should boot from by default. If next is used, set the boot location to be the next one after the one currently booted from. (This does not mean the next one after the one we are currently set to boot from; thus the command is idempotent, and will not continue to cycle through all of the available locations.)

Control Boot Manager Password boot bootmgr password boot bootmgr password 0 535-0008-00 Rev 10


309


boot bootmgr password 7 no boot bootmgr password Configure or remove a password to control access to boot manager parameters. Similar to the username * password ... commands, it is generally intended to be used in its cleartext forms; the form that takes an encrypted password mainly exists so that show configuration will have something to display.

Show Images show images Show all image files on the system, as well as what images are installed in two locations: the active location (which was most recently booted from) and the default location (which is the default to boot from in the future). There may or may not be overlap between these two lists. show bootvar Similar to show images in that it displays what images are on the two locations, and which are the active and default location. But not all of the show images data is displayed, and additionally this displays whether or not a boot manager password is set.

Enable and Config Modes enable Enter enable mode, if the user account was set up with the privileges to do so. disable Leave enable mode. configure terminal conf t Enter Config mode, if the current user account has the privileges to do so. Note that there is no mechanism to prevent multiple users from being in Config mode simultaneously. no configure Exit Config mode and return to enable mode. exit Exit the current mode. From Config mode, go to enable mode. From enable or standard mode, log out of the system. Note that to go from enable mode to standard mode, the disable command must be used. help Display a general CLI help message.

310


535-0008-00 Rev 10


Web Proxy Settings This section provides the syntax for web proxy setting commands, along with an explanation of the functionality of each command. web proxy host [port ] no web proxy If desired on a http or ftp download, a proxy can be specified. If no port is specified, the default is 1080.

Xinetd Server Settings This section provides the syntax for xinetd server setting commands, along with an explanation of the functionality of each command.

Enable and Disable a Telnet Server [no] telnet-server enable Enable/disable the telnet server.

Show Telnet Server Settings show telnet-server Show telnet server settings.

Enable and Disable an FTP Server [no] ftp-server enable Enable/disable the FTP server.

Show FTP Server Settings show ftp-server Show FTP server settings.

tcpdump Capture Files This section provides the syntax for tcpdump commands, along with an explanation of the functionality of each command.

Display tcpdump Files show files tcpdump Display a list of tcpdump capture files.

535-0008-00 Rev 10


311


Delete a tcpdump File file tcpdump delete Delete the specified tcpdump output file.

Upload a tcpdump File file tcpdump upload Upload the specified tcpdump output file to the specified URL. Only scp pseudo-URLs are supported for the destination. See Key to Command Parameters on page 288 for the scp:// URL format.

NTB Commands (vMOS Data Management) The Non-Transparent Bridge (NTB) is automatically enabled during installation. If this system is running vMOS-6, you can use the commands in this section verify that NTB is enabled and to show the status of the NTB link. Verifying that NTB is Configured Perform the following command in enable mode. show ntb configured violin-vmga [violin-vmg: master] # show ntb configured vntb0 : Enabled: yes MTU: 16320 Mask Len: 30 mg-a IP: 169.254.2.1 mg-b IP: 169.254.2.2 Verifying that the NTB Link is Up on Both Gateways Perform the following command in enable mode. show ntb status violin-vmga [violin-vmg: master] # show ntb status vntb0 link status : Link Local Remote ------------ ---------PCI (local) : Up Up PCI (peer) : Up Up NTB : Up Up Network : Up Up

312


535-0008-00 Rev 10


Cluster Configuration and Show Commands Show Cluster Commands show cluster configured Display global cluster configuration state. show cluster local Display local cluster run state. show cluster local error-status Display error status of local node. show cluster master Display run state information about master node. show cluster global Displays global cluster run state. show cluster standby Display run state information about standby node. Array and Cluster Action Commands array format capacity Format the Memory Array to a specific capacity. array modules Configure a Memory Array module. array reboot [acms-only | mgs-only | vcms-only] Reboot all Array Controller Modules, internal Memory Gateways or vRAID Controller Modules. array shutdown Shut down the following modules in the Violin 6000 Series Memory Array: Array Controller Modules, vRAID Controller Modules, internal Memory Gateways, VIMMs and Host Bus Adapters. This command also turns off the LEDs on the front of the Violin 6000 Series Memory Array. array upgrade modules type vcm Upgrade all VCMs. array upgrade modules id vcm-x Upgrade a particular VCM, where vcm-x is vcm-a, vcm-b, vcm-c or vcm-d.

535-0008-00 Rev 10


313


system upgrade Upgrade all nodes with the cluster to the specified image. This must be performed on the Master. Cluster Configuration Commands cluster master address vip Set the cluster master virtual IP address and netmask. no cluster master address vip Reset cluster master virtual address to default. Display Statistics Configuration and Gathered Data show stats alarm [alarm id] Display status and configuration of statistics-based alarms. show stats chd Display configuration of statistics CHDs. show stats cpu Display CPU statistics. show stats media Display media statistics. show stats sample Display configuration of statistics samples.

Network Configuration Commands This section describes the CLI commands for network configuration. In general, the commands described in this section only work when you are in Config mode.

• • •

314

Network Bond Commands on page 315 Name Resolution Commands on page 315 Routing Commands on page 316


535-0008-00 Rev 10

Network Configuration Commands

Network Bond Commands The network bond command enables you to define two or more network interfaces (data ports) as a single logical address called eth0 and to define the mode by which these interfaces are aggregated together. network bond eth0 mode [balance-abalance-rr|backup|link-agg|link-agglayer2+3|link-agg-layer3+4] interface interface The command aggregates two interfaces together as a bonded group called eth0. The interfaces share the same subnet.

Ten network bonding modes are supported: balance-rr

Round robin balancing

backup

Backup fault-tolerant mode

link-agg

Link Aggregation mode

link-agglayer2+3 link-agglayer3+4

Link Aggregation Layer 2 + 3 mode Link Aggregation Layer 3 + 4 mode

Name Resolution Commands hostname no hostname Set or clear the system hostname. [no] ip name-server [no] ip domain-list [no] ip host The commands configure DNS servers, specifies which domain(s) to try unqualified hostnames in, and defines hostname/IP mappings for /ETC/HOSTS. show hosts Show all of the following values: hostname, name servers, domain name list, and static host mappings.

535-0008-00 Rev 10


315


Routing Commands ip default-gateway no ip default-gateway Set or remove the default route. ip route no ip route [] Set or remove a static route. If it is called with only a network prefix and mask, it deletes all routes for that prefix. show ip route Display the routing table in the system, which includes dynamic routes as well as any active static routes. show ip route static Display the list of configured static routes show ip default-gateway [static] Display the currently active default route (or the configured one, if static is used). This is redundant with the show ip route [static] command, but is provided to enhance usability.

Administrative Commands This section describes the administrative commands. In general, the commands described in this section only work when you are in Config mode.

donotuse Command VIMMs and VCMs that are powered off or that have failed are flagged by the system as a module that should not be used or may need to be replaced. This flag—which can be displayed in the CLI by typing “show donotuse”—is a reminder to the system administrator if a VIMM or VCM fails. This flag may also display for VIMMs and VCMs that do not require replacement. show donotuse The show donotuse command shows VIMMs and VCMs that are flagged as donotuse faults, including user- and system-set faults. The following is an example of the show donotuse command along with output for a system-set fault: # show donotuse vimmid system VIMM is not responding at 11-01-2012 20:41:11 UTC█ User-set faults enable the donotuse flag when an administrator issues one of the following commands:

• 316

no array modules id vimmid | vcm-x enable Violin 6000 Series Memory Array User’s Guide

535-0008-00 Rev 10

Email Alert Consolidation Commands

•

no array modules type vcm enable

The following is an example of the show donotuse command along with output for a user-set fault: # show donotuse vimmid user-set VIMM is not responding at 11-01-2012 20:41:11 UTC█ no array modules id vimmid donotuse Use the above command to clear the donotuse flag for a VIMM, where id is the VIMM number. no array modules id vcm-x donotuse Use the above command to clear the donotuse flag for a VCM, where x is the VCM letter (a, b, c or d). no array modules id donotuse Use the above command to clear the flag for all system-set faults, where component is either VIMM or VCM. no array modules id enable Use the above command to clear the flag for all user-set faults, where component is either VIMM or VCM.

Email Alert Consolidation Commands You can manage email alerts, reducing the number of emails received, with the email consolidation feature. Important points to remember about email consolidation:

•

If a system begins shutdown or if a maximum number of emails are staged (default 5), they are sent out automatically.

•

Emails are consolidated on a per-host basis. Emails from different hosts (acm-a, acm-b, mg-a, or mg-b) cannot be combined.

•

Log in to the ACM master or MG master node to configure email consolidation. The default is disabled.

Enable and Disable Email Consolidation You can enable and disable email consolidation using the following command. The default is disabled. Log in to the desired host (acm-a, acm-b, mg-a, mg-b), then issue the command.

Enabling Email Consolidation email consolidate Turns on email consolidation. The default is disabled.

535-0008-00 Rev 10


317


Disabling Email Consolidation no email consolidate Turns off email consolidation.

Customize Email Consolidation When email consolidation is enabled, the default settings are as follows:

• •

Time period before email is sent: 60 seconds Maximum number of events that can be combined: 5

Setting the Wait Period email consolidate period [time] Sets the wait time in seconds, before the email is sent.

Setting the Number of Combined Events email consolidate events [num] Sets the maximum number of combined events.

View Current Email Settings You can view the current email consolidation settings with the following command. show email Output similar to the following is shown. vmem-acm-b [system1: master] # show email Mail hub: Mail hub port: 25 Domain: (system domain name) Return address: do-not-reply Include hostname in return address: yes ... Consolidate email settings: Enable: no Period: 60 Max_events: 5 ...

318


535-0008-00 Rev 10

Unified Logging Upload Commands

Unified Logging Upload Commands This section covers the unified logging upload commands and their respective purposes.

Upload Log Files for All Nodes logging files upload-auto immediate When run from the internal Memory Gateway master, /var/log/messages are collected from both Memory Array nodes (ACM) and both internal Memory Gateway nodes (MG), and then uploaded to a pre-configured target.

Cancel a Log File Upload Process logging files upload-auto immediate cancel Stops an in-process log file upload.

Upload Log Files From the Local Host logging files upload-auto immediate local Only uploads /var/log/messages from the local host.

Upload Log Files From a Specific Node logging files upload-auto immediate module id Uploads log files from a specified Node (acm-a/b, mg-a/b).

Upload Log Files From a Specific Node Type logging files upload-auto immediate module type Uploads log files from a specific node type (acm or mg).

Address Resolution Protocol Commands The following commands allow you to configure ARP, a protocol used for resolution of network layer addresses into link layer addresses. This is a critical function in multiple-access networks.

Add and Remove Static ARP Entries arp Add a static entry to ARP. no arp

535-0008-00 Rev 10


319


Remove a static entry from ARP. Note that this cannot be used to remove dynamic entries.

Clear Dynamic ARP Entries clear arp-cache Clear dynamic entries in the ARP cache. Note that this does not delete static ARP entries configured with the arp ... command.

Display Contents of ARP show arp Display the contents of ARP. This should contain all of the statically-configured ARP entries, as well as any that the system has picked up at dynamically.

List Statically-Configured ARP Entries show arp static Display a list of all statically-configured ARP entries.

Array Configuration and Show Commands You can use the array set of commands in the CLI to configure and monitor the Array Controller Modules (ACMs).

Array Configuration Commands The following array specific commands allow you to format the array, power up or power down a specific array module or ID, configure a PCIe port/profile, and display array-related information and system alarms log. The Violin 6000 supports four vRAID Controllers. You can format all four of the vRAID Controllers’ storage capacity into one large array entity by using the array format capacity command at the configure terminal sub-mode. An example of the array format capacity command is shown below.

Caution: The Memory Array is preformatted to specific storage capacities, depending on the type of VIMMs in the system. A single level cell (SLC) system is formatted to 65% and a multi-level cell system is formatted to 84%. For optimum system performance, do not change these values.

The following storage capacity percentages are supported: 50%, 65%, 78%, 84% and 87%. violin-acma [violin: master] (config) # array format capacity █ 320


535-0008-00 Rev 10

Array Configuration and Show Commands

Use the array modules type enable command to power up a specific array module type: violin-acma [violin: master] (config) # array modules type enable█ Use the [no] prefix to power down a specific module type, for example: violin-acma [violin: master] (config) # no array modules type vcm enable█ Use the array cooling command to adjust the allowable temperature of the Memory Array. This command reduces the speed of the fans based on the preferred setting. violin-acma [violin: master] (config) # array cooling █ The eco option allows the Memory Array to run at a higher temperature before the fans increase in speed to lower the temperature of the system.

Array Show Commands Use the show array set of commands to view details about the array. The following example shows an array that contains 64 VIMMs, has no alarms, a controller temperature of 28 degrees centigrade and an ambient temperature of 24C: violin-acma [violin: master] # show array Array Type #VIMM Alrm Ready T/Con T/Amb MGs Status ------------------------------------------------------------------------------------VIOLIN_MEMORY_ARRAY_23109R00000012 V6000 64 yes no 24 22 optimal violin-acma [violin: master] #█

535-0008-00 Rev 10


321


Use the show array detail command to show more information about the array. The following example includes information about power supplies, LEDs and fans: violin-acma [violin: master] # show array detail ARRAY: VIOLIN_MEMORY_ARRAY_23109R00000012 INDEX: 0 Chassis Type : V6000 Number of VIMMs : 64 Ambient Temp : 22 Controller Temp : 24 Power A : OFF Power B : ON Lid Ajar Time : 0 secs Alarm LED : ON Power A LED : OFF Power B LED : ON MG-a Status : running MG-b Status : running Fans: Fan a1 : High Fan b1 : Slow Fan c1 : Slow Fan a0 : High Fan b0 : Slow Fan c0 : Slow violin-acma [violin: master] #█ show array modules Display information about all of the array modules. show array summary Display summary information for arrays. Additional show commands are available by typing show ? at the command prompt. show array modules type Display information about a module type. See General Configuration Commands on page 292 for additional Array configuration commands, and the Violin 6000 Series Memory Installation Guide for the initial steps to configure the Violin 6000 Series Memory Array.

Array Balance Commands RAID groups may become unbalanced over time if VIMMs are replaced and if other administratorinitiated changes are made. If two or more VIMMs in a RAID group share the same root VIMM, the RAID group is considered “unbalanced”, causing the system to perform less than optimally. Array balance is disabled by default.

322


535-0008-00 Rev 10

Array Configuration and Show Commands

Balance of the system can be restored by “moving” VIMMs from RAID rebuilds. If necessary, multiple RAID rebuilds can be performed at the same time. array balance enable The above command enables automatic balancing. When unbalanced RAID groups are detected, they are automatically rebuilt. array balance The above command starts a RAID rebuild for each VCM that requires a rebuild. More than one rebuild may be required to balance the system. no array balance enable Use this command to disable (turn off) automatic rebalancing. array balance schedule once Use the above command to schedule an array balance for the default time of the upcoming Saturday at 1:00 A.M. This starts a RAID rebuild for each VCM that requires a rebuild. More than one rebuild may be required to balance the system. no array balance schedule once Use the above command to cancel a scheduled one-time rebalance. show array balance Use the above command to display the current RAID rebalance settings and view the status of all the VCMs. array balance schedule once [day-of-week ] [time ] Use the above command and parameters to set a one-time RAID rebalance at a specified time and day of the week. array balance schedule weekly [day-of-week ] [time ] Use the above command and parameters to set a specific time and day of the week on a recurring basis. Multiple RAID rebuilds are completed until the system is balanced. The scheduled array balance is disabled once the system is in balance. no array balance schedule weekly Use the above command to cancel scheduled weekly recurring rebalances. See Rebalancing RAID Groups on page 242 for more information on using the array balance commands.

535-0008-00 Rev 10


323


VIMM Commands This section describes the VIMM configuration and monitoring commands that you can use in the CLI. Use the array modules id vimm command to power on VIMMs. Use the [no] prefix to power off a VIMM. For example: violin-acma [violin: master] (config) # [no] array modules id vimm enable█ Where: [no]

Specifies that a VIMM is to be disabled.

id

Enable or disable a specific VIMM.

The specific VIMM number to disable.

Use the show vimms set of commands to display all or specific VIMM information. For example: violin-acma [violin: master] (config) #show vimms [detail | summary]█ Where:

324

Shows detailed information about all or a specific VIMM.

Shows a summary of the VIMMs.


535-0008-00 Rev 10

iSCSI Commands

The following example shows output of the show vimms detail command for one VIMM: violin-acma [violin: master] (config) #show vimms detail VIMM01: VCM : vcm-a Type : 512G-MLC-Flash Status : Active Present : yes Power : yes DoNotUse Fault : no Current (mA) : 989.38 RAID : 1 Spare : no Health : health threshold: (OK) Temp (C) : 39 Serial : 2610CN00002723 Model : 620-0046-01_R03 Date : 20110212 FpgaVersion : 19935 SwVersion : 19892 %-Format Capacity : 65.00 %-DieFail : 0.00 %-BlkFail : 0.00 %-BlkEraseAvg : 2.00 %-LifeTimeRemaining : 98.00█ vvimms Use this command to display the Violin 6000 Series Memory Array VIMMs.

iSCSI Commands This section provides the syntax for iSCSI commands, along with an explanation of the functionality of each command.

Show iSCSI Targets show iscsi The show iscsi command shows if iSCSI has been enabled, the number of bindings, the IP address, and interface for each binding. (config) # show iscsi iSCSI is enabled iSCSI Target: iscsi

#Bindings: 4

10.1.9.82 (unknown)

535-0008-00 Rev 10


325


192.168.1.101 (eth3) 192.168.2.100 (unknown) 192.168.3.101 (eth4)

Enable iSCSI Target Ports [no] iscsi enable The iscsi enable command may be used to enable or disable iSCSI target ports.

Create an iSCSI Target [no] iscsi target create The iscsi target create command enables you to create and name a target. Additional target names may be created if needed. The target name consists of up to 32 alphanumeric characters. A user-defined name for the target. The term iSCSI is the default name used for the target automatically created when you use the iscsi enable command.

Bind an iSCSI Target [no] iscsi target bind name to [to ]* The iscsi target bind command enables you to bind an iSCSI target to one or more IP addresses. The command limits visibility of the target to initiators which have access to those IP addresses.

Security Commands This section covers the following commands:

• • •

iSCSI CHAP Configuration on page 326 RADIUS Configuration on page 328 TACACS+ Configuration on page 329

iSCSI CHAP Configuration Enforces authenticated sign-on to ensure that Internet Small Computer System Interface (iSCSI) initiators and targets prove their identity to each other using the Challenge-Handshake Authentication Protocol (CHAP). For more information, see iSCSI Authenticated Sign-On on page 248.

326


535-0008-00 Rev 10

Security Commands

CHAP Requirements iSCSI CHAP configurations must meet the following requirements:

• • •

The minimum CHAP secret size is 12 characters (bytes) (IPSec is not supported).

•

For bidirectional CHAP, each target-initiator pair must have one target CHAP secret and one initiator CHAP secret configured.

•

In bidirectional CHAP mode, the target CHAP secret cannot be identical to the initiator’s CHAP secret to which it is connected.

The maximum CHAP secret size is 64 characters (bytes). For unidirectional CHAP, each target-initiator pair must have one target CHAP secret configured.

Note: iSCSI CHAP configurations are persistent across system reboots.

Configuring Unidirectional CHAP iscsi chap target {target-secret} to {initiator |igroup } Sets the CHAP target secret for a target-initiator, or target-initiator group, pair. The igroup_name variable option, as used in this command, allows you to establish the CHAP secret for a group of initiators all at once. However, when you view iSCSI CHAP configurations, as described in Showing iSCSI CHAP Configurations on page 328, the individual initiators that are members of the igroup are shown individually without reference to the igroup_name (they belong to).

Configuring Bidirectional CHAP iscsi chap target {target-secret|initiator-secret} to {initiator |igroup } Sets the CHAP target secret and initiator, or initiator group, secret. The igroup_name variable option, as used in this command, allows you to establish the CHAP secret for a group of initiators all at once. However, when you view iSCSI CHAP configurations, as described in Showing iSCSI CHAP Configurations on page 328, the individual initiators that are members of the igroup are shown individually without reference to the igroup_name (they belong to).

Removing CHAP Secrets no iscsi chap target to {initiator |igroup } Removes the CHAP target secret and initiator secret for a target.

535-0008-00 Rev 10


327


Showing iSCSI CHAP Configurations Use the show iscsi chap command to view target and initiator CHAP configurations, as shown in the following example. (config) # show iscsi chap Target Initiator Target Initiator Chap Chap ------------------------------------------------------------------Yes Yes iscsi iqn.1991-05.com.microsoft:win2k8-lt Yes Yes iscsi iqn.1991-05.com.microsoft:violin-w2k3r2 Yes No iscsi-2 iqn.1994-05.com.redhat:981b2975f144

RADIUS Configuration radius-server timeout no radius-server timeout Sets (or resets to the default) a global communication value for all RADIUS servers. Can be overridden in a radius-server host command. The default is 3. Sets the timeout for retransmitting a request to any RADIUS server. Range is 1-60. radius-server retransmit no radius-server retransmit Sets (or resets to 0) a global communication value for all RADIUS servers. Can be overridden in a radius-server host command. Defaults to 1. Sets the number of times the client will attempt to authenticate with any RADIUS server. To disable retransmissions set it to zero. Range is 0-5. radius-server key no radius-server key Sets (or clears) a global communication value for all RADIUS servers. Can be overridden in a radius-server host command. Sets the shared secret text string used to communicate with any RADIUS server. radius-server host {hostname , ip-address} [auth-port ] [timeout ] [retransmit ] [key ] no radius-server host {hostname , ip-address} [auth-port ] Add a RADIUS server to the set of servers used for authentication. Some of the parameters given may override the configured global defaults for all RADIUS servers. The auth-port defaults to 1812 and is used for authentication requests. The same IP address can be used in more than one radius-server host command as long as the auth-port is different for each. auth-port is a UDP port number. auth-port must be specified immediately after the host option (if present). If no radius-server host {hostname , ip-address} is specified, all radius specific configuration for this host is deleted. no radius-server host {hostname , ip-address}

328


535-0008-00 Rev 10

Security Commands

auth-port may be specified to refine which host is deleted, as the previous command will delete all RADIUS servers with the specified ip-address. RADIUS servers are tried in the order they are configured. show radius Show the RADIUS configuration.

Note: We do not have CLI commands to specify the acct-port (accounting port) or retransmit on a per server basis. Also, there is no radius-server deadtime command as there is in the Cisco command set.

TACACS+ Configuration This section provides the syntax for TACACS+ configuration commands, along with an explanation of the functionality of each command.

Set Wait Time for TACACS+ Server tacacs-server timeout no tacacs-server timeout Sets (or resets to the default) a global communication value for all TACACS+ servers. Can be overridden in a tacacs-server host command. Defaults to 5. Sets the wait time for retransmitting a request to any TACACS+ server. Range is 1-60.

Set Client Authentication Attempts tacacs-server retransmit no tacacs-server retransmit Sets (or resets to 0) a global communication value for all TACACS+ servers. Can be overridden in a tacacs-server host command. Defaults to 2. Sets the number of times the client will attempt to authenticate with any TACACS+ server. To disable retransmissions set it to zero. (Number of times to search the TACACS list.) The range is 0-5.

Set Shared Secret tacacs-server key no tacacs-server key Sets (or clears) a global communication value for all TACACS+ servers. Can be overridden in a tacacs-server host command. Sets the shared secret text string used to communicate with any TACACS+ server.

535-0008-00 Rev 10


329


Add a TACACS+ Authentication Server [no] tacacs-server host {hostname , ip-address} [auth-port ][auth-type {ascii, pap}] [timeout ] [retransmit ][key ] Add a TACACS+ server to the set of servers used for authentication. Some of the parameters given may override the configured global defaults for all TACACS+ servers. The auth-port is used for authentication requests. The auth-type specifies which of the two currently supported authentication methods will be used. The same IP address can be used in more than one tacacs-server host command as long as the auth-port is different for each. auth-port is a UDP port number. auth-port must be specified immediately after the host option (if present). If no tacacs-server host {hostname , ip-address} is specified, all tacacs-specific configuration for this host is deleted. no tacacs-server host {hostname , ip-address} auth-port may be specified to refine which host is deleted, as the previous command will delete all TACACS+ servers with the specified ip-address. TACACS+ servers are tried in the order they are configured.

Show the TACACS+ Configuration show tacacs Show the TACACS+ configuration. Currently there is no “single-connection” option on a per TACACS+ server basis.

SCSI-3 ALUA Commands Asymmetric Logical Unit Access (ALUA) is a multipathing technology that enhances availability through efficient data access. This is done through the correct handling of path failures with out relying on client timeout mechanisms.

Note: By default ALUA is not enabled. To use this feature, you must enable ALUA on a per LUN

basis. lun set container name alua Enables ALUA for a single LUN in a specified container. lun set container name * alua Enables ALUA for all LUNs in the container through the use of the wildcard (*) character. lun create container name size alua Enables ALUA for the newly created LUN.

330


535-0008-00 Rev 10

Fibre Channel Commands

Fibre Channel Commands You can use the following CLI command to set the Fibre Channel port parameters, such as Speed Option, Port Type, enabling and disabling a port, as well as resetting the Fibre Channel link. fc {on | target} {global | hostname} {all | hostnamexxx} target {all | hba-xx} | enable | issue_lip | speed {auto, 2Gb, 4Gb, 8Gb} | type {loop, loop-preferred, point-to-point} To reset the port settings to default value, use the following command. no fc {on | target} {global | hostname} {all | hostnamexxx} target {hbaxx | all} {enable | speed | type}

• • •

speed: Default is “auto” type: Default is “point-to-point” enable: Default is enable

Initiator Group Commands This section covers the following commands:

• • • • • •

Create an Initiator Group on page 332 Add Initiators to a Group on page 332 Create a Nested Initiator Group on page 332 Show Initiator Groups on page 332 Show Nested Initiator Groups on page 333 Show Initiators on Targets on page 333

This section provides the syntax for initiator group and initiator group related commands. An explanation of the functionality of each command is also provided.

•

Changing the LUN ID after a LUN has been exported is not recommended. For example, if a LUN is exported as LUN ID 1 and later the export is changed to LUN ID 2, traffic is disrupted during the change. The affect on the client depends on how it handles LUN ID changes.

•

You can export LUNs to different igroups on different targets, even when they have initiators in common.



535-0008-00 Rev 10


331


Create an Initiator Group [no] igroup create name The igroup create command creates an initiator group, after which you can add initiators using the addto command. For more information, see Add Initiators to a Group on page 332.

Add Initiators to a Group [no] igroup addto name initiators [initiator_name …] The igroup addto command adds one or more iSCSI, Fibre Channel, or InfiniBand initiators. iSCSI initiator identifiers (IQNs) are user-defined. Fibre Channel initiator identifiers (WWNs) are generated automatically by a Host Bus Adapter (HBA). InfiniBand initiator identifiers begin with a guid prefix.

Create a Nested Initiator Group Best practices for nested igroups:

• •

Use a unique LUN ID (LUN name) for each LUN to avoid the chance of conflicts.

•

Refrain from exporting a LUN to different targets using a different LUN ID (LUN name) for the LUN for each target, as this practice can result in conflicts.

Plan the nested igroups hierarchy first. Then, export the LUNs to the different igroups, allowing the system to auto-assign LUN IDs to ensure they are each unique.

[no] igroup addto name initiators The igroup addto command allows you to include an igroup as a member of another igroup. The nested igroup becomes the initiator in this instance. Nested igroups can contain iSCSI, Fibre Channel, and InfiniBand initiators. In the following example two igroups are created: A and B. The B igroup is then added to A, and A becomes a nested igroup. igroup igroup igroup igroup

create name A create name B addto name A initiators B addto name A initiators wwn.50:01:43:80:18:6b:3e:f2

Now the igroup A contains two members wwn.50:01:43:80:18:6b:3e:f2 and B. Since B is an igroup, we call the A is a nested igroup.

Show Initiator Groups show igroups Displays all the initiator groups with their members.

332


535-0008-00 Rev 10

LUN Commands

Show Nested Initiator Groups show igroups flatten Displays all the initiator groups along with their members, replacing initiator group members with initiators. This shows the nested igroup data by flattening the nested hierarchies.

Show Initiators on Targets show targets initiators Displays a list of initiators with their associated targets. show targets initiators no-igroup Displays a list of initiators that are not currently members of an igroup, along with their associated targets.

LUN Commands This section covers the following commands:

• • • • • • •

Create LUNs on page 333 View LUNs on page 334 Set LUN Privileges and UDIDs on page 335 Rename a LUN on page 335 Resize a LUN on page 336 Export LUNs on page 336 View LUN Statistics on page 337

This section provides the syntax for LUN commands, along with an explanation of the functionality of each command. VMware ESX 4.1 does not support 4K block size. For VMware ESX 4.1, use a 512 byte block size LUN, which is supported.

Create LUNs The lun create command enables you to create a LUN within a specified storage container. Command Syntax

[no] lun create container name size [ | equal] [quantity ] [nozero] [readonly] [startnum ] [blksize 512 | 4096] [offline] alua naca

535-0008-00 Rev 10


333



The table shows syntax for LUN commands. name

Defines the name of LUN to be created; used as a prefix if the quantity of LUNs created is greater than one.

size

Defines the size of each LUN in GB, or “equal” to divide the free space of the container equally among the new LUNs. A value of 0 (zero) may be used in place of “equal.”

container

Identifies the name of the container in which the LUN is created. Defines the number of LUNs to be created; the default is 1. If the number is greater than 1, the names of the newly created LUNs will be appended with an index number from startnum up to the number of LUNs created.

quantity

nozero

Specify to avoid zeroing-out the data on the LUNs, which could take a long time for large LUNs.

readonly

Creates the LUNs in read-only mode.

startnum

Defines starting index number for newly created LUNs. The default is 1. Defines the logical block size for the created LUNs; the default is 512, can be set to 512 or 4096. Not all systems can handle 4096byte sizes. Use 512 if you are in doubt.

blksize

offline

Creates the LUNs in offline mode.

alua

Turns on ALUA for a LUN. For more information on ALUA, see Optimizing Performance with ALUA on page 257.

naca

Turns on NACA for a LUN. For more information on NACA, see AIX Compatibility NACA Bit Support on page 179.

View LUNs The show luns command enables you to view LUNs. LUNs may be filtered by container, name, and session parameters. To view a list of containers, enter show luns container ?. The name parameter enables you to view only those LUNs which are prefixed by a particular name. The udid option shows the UDID associated with a LUN. show luns [container] [name] [count] [reserved] [serial] [sessions] The following table provides descriptions for each command argument and option.

334


535-0008-00 Rev 10

LUN Commands


container

Displays only the LUNs within a specific container.

name

Displays a specific LUN.

count

Displays LUNs with active sessions and their count.

reserved

Displays LUNs with SCSI reservations.

serial

Displays the serial numbers for each LUN.

sessions

Displays the sessions accessing a particular LUN.

Set LUN Privileges and UDIDs [no] lun set container name readonly The lun set command enables you to set a LUN as read-only. The lun set command is the only command usable after a LUN has been created except the no lun create … readonly command. # lun set container name udid Sets the unique device identifier (UDID) for the LUN, after which you are prompted to confirm the change. # lun set container name * udid Sets the UDIDs for all LUNs in the container, starting at 100 and increasing incrementally for each additional LUN (101, 102, 103 ...). The * acts as a wildcard. # no lun set container name * udid Removes the UDIDs for all LUNs in a container. The * acts as a wildcard.

Rename a LUN You can rename a LUN using the following command.

Note: Renaming or deleting a LUN is not allowed with actively exported LUNs. You must

unexport the LUN before you renaming or deleting it.

[no]lun rename container name to

535-0008-00 Rev 10


335


Resize a LUN You can increase the size of an existing LUN by specifying a new size or specifying the amount you want to increase the existing size in gigabytes. Decreasing the size of an existing LUN is not supported. For a smaller size LUN, create a new LUN of the desired size. lun resize container name size Increases the size of an existing LUN by specifying the new size in gigabytes. lun resize container name size + Increase the size of an existing LUN by specifying the amount (GBs) to increase the current size.

WARNING! Existing client side file systems that might be stored on the LUN are not extended as part of the resize operation. Increasing the size of an existing client side file system must be done on the client. Some client systems do not detect a LUN size change until they reboot.

Export LUNs The lun export command enables you to export the LUNs within a specific container to an initiator or initiator group. A single LUN may be exported multiple times through different target ports (multipathing) or to different initiators (clustered access).

Caution: When working with initiator groups or nested initiator groups, review the recommended

guidelines as described in Exporting LUNs on page 193 prior to exporting.

Syntax

[no] lun export container name [lunid][to ]... [using ]

336


535-0008-00 Rev 10

LUN Commands


lun_name

The lun_name could be wildcard specified, such as FINANCE*

container

The name of the container

name

The name of the LUN

lunid

Designates the LUN ID assigned to an exported LUN. By default, LUN IDs are automatically assigned by vSHARE. User-defined LUN IDs may be specified to track specific LUNs.

igroup

The name of the igroup. If no igroup or initiator is specified, all initiators will have access to the LUN.

initiator

The IQN (iSCSI) or WWN (Fibre Channel) name of the initiator.

port

The port is iSCSI portal (i.e. 10.10.0.14) or Fibre Channel wwn ID.

View LUN Statistics LUN statistics are accessed via the CLI with the following command: show stats lun [container ] [name ] [on global|on host ] [continuous] [detail] Use the container option to restrict statistics display to a single container. If it is omitted, statistics are shown for all the containers accessible to the hosts. Use the name option to get statistics for a single LUN. If it is committed, statistics for all LUNs are specified.

Note: Select shell wildcard patterns can be used in the field to match a set of LUNs. For example, you can use the asterisk (*) to match all characters that follow. Such as mylun* to match all LUN names that begin with “mylun”. Or, you can specify a range of LUN names using the [a-e] wildcard. Such as my-lun[2-4][0-9] to match all LUNs from my-lun20 to my-lun49.

Use the on global option to get statistics for data flowing to LUNs through all nodes in a cluster. Use the on host option to get statistics for data flowing to LUNs on a specific host. Use the continuous option to refresh displayed statistics every ten seconds. Use the detail option to display statistics in a verbose format.

535-0008-00 Rev 10


337


LUN Group Commands This section covers the following commands:

• • • • •

Creating LUN Groups on page 338 Deleting LUN Groups on page 338 Modifying LUN Groups on page 338 Exporting a LUN Group on page 339 Displaying LUN Groups on page 340

Creating LUN Groups The following guidelines apply:

• • •

A LUN may be part of multiple LUN groups.

•

A LUN group name must be unique within a container.

A LUN group can contain a minimum of one LUN and a maximum of eight LUNs. A LUN group name must start with an alphanumeric character, and may contain hyphens, underscores, and periods.

Substitute LUN names for the variables in the command above. You must have one LUN to create a LUN group, and optionally can specify seven other LUNs as members. # lungroup create container name lun lun [ lun lun ... lun ]

Deleting LUN Groups You can delete a LUN group without affecting the LUNs belonging to the group. Deleting a LUN group in effect only deletes the relationship that binds the member LUNs together as an object.

Note: You must delete the snap groups (snapshots of a LUN group) associated with a LUN group before you are allowed to delete the LUN group. Snapshots that belong to a group cannot be deleted individually. You must delete all the snapshots in a snap group.

# no lungroup create container name

Modifying LUN Groups You can change a LUN group name, as well as the LUNs that are members of the group. You can add new LUNs to a LUN group, and remove existing LUN members. This function only impacts the management of the LUN group (object), and does not create or delete actual LUNs.

Note: If a LUN is deleted, it is automatically removed from any LUN groups of which it is a member. If deleting a LUN results in a LUN group with zero members, then that LUN group is automatically deleted.

338


535-0008-00 Rev 10

LUN Group Commands

Adding LUNs to a LUN Group Substitute appropriate names for the variables in the following command. # lungroup addto container name lun Note: A LUN group can have a maximum of eight LUNs.

Deleting LUNs from a LUN Group Substitute LUN names for the variables in the following command. # no lungroup addto container name lun lun [ lun lun ... lun ] Note: You can delete up to seven LUNs from a LUN group. One LUN must remain to maintain

the group. If you delete all LUNs from a LUN group, the LUN group is automatically deleted.

Renaming a LUN Group To rename a LUN group, specify the container in which the LUN resides, the existing name (old name) of the LUN followed by the new name of the LUN.

Note: Renaming a LUN group with exported LUNs is not recommended, as traffic can be

disrupted. For more information, see Guidelines for Exporting LUNs on page 194. # lungroup rename container name to

Exporting a LUN Group You can export a LUN group and all the LUNs contained within the LUN group, using the following options. — to: exports a LUN group to an initiator or initiator group — using: exports a LUN group using a target port — description: allows you to add a description to identify the LUN group Rules for exporting LUNs also apply to LUN groups. For more information, see Guidelines for Exporting LUNs on page 194.

Exporting a LUN Group to an Initiator Port The following command exports a LUN group to a specified initiator port. # lungroup export container name to

535-0008-00 Rev 10


339


Exporting a LUN Group Using Specific Ports The following command exports a LUN group using specified ports. # lungroup export container name using



Displaying LUN Groups You can display a list of all LUN groups on a system. Each LUN group name and the LUNs present in each group are shown. When a LUN name changes, the LUN group list shows the new name. You can specify a container, and only show the LUN groups in that container. The following command displays a list of LUN groups and their individual members. # show lungroups [container ]

Container: MyContainer LUN Group LUNs -------------------------------------------------------------Finance_LG 2013,1stQ,2ndQ,3rdQ,4thQ HR_LG Benefits,Employees,Expenses RandD_LG Resources,Expenses,HeadCount Eng_thn_LG eng1_thn,eng2_thn,eng3_thn,eng5_thn Admin_thn_LG adm1_thn,adm2_thn,adm3_thn Corp_thn_LG corp5_thn,corp6_thn,corp7_thn

340


535-0008-00 Rev 10

Data Management Commands

Data Management Commands This section covers the commands for the following:

• • • • • •

Optimizing Performance with ALUA on page 341 Thin Provision LUNs on page 344 Snapshots on page 346 Snap Groups on page 350 Schedule Management on page 353 Free Space Management on page 356

Optimizing Performance with ALUA Implicit Asymmetric Active-Active (ALUA) directs management traffic to one preferred port, to increase performance for thin provisioned LUNs and LUNs with snapshots. ALUA is achieved by setting a preferred port attribute and an ALUA attribute on a per-LUN basis.


a substantial performance hit.

There is a preferredport LUN attribute that is used in conjunction with the alua LUN attribute that sets the preferred bit and target port asymmetric access state of each target port group.

Setting the Preferred Port and ALUA Attributes for a New LUN You can set both the preferredport and the alua attributes when creating a new LUN. Optionally, you can set the preferred port for a LUN first (as described in Setting a Preferred Port for an Existing LUN on page 341), and then set the alua attribute at a later time (as described in Setting ALUA for an Existing LUN on page 342).

Note: The preferredport setting is only applicable when the alua attribute is enabled on a

LUN. If a preferred port is not specified, it defaults to 1. # lun create container name size [preferredport ] alua

Setting a Preferred Port for an Existing LUN You can set the preferredport attribute on an existing LUN.

Note: The preferredport setting is only applicable when the alua attribute is enabled on a

LUN. If the alua attribute is not enabled, the preferredport setting has no effect. # lun set container name preferredport

535-0008-00 Rev 10


341


Setting ALUA for an Existing LUN This section demonstrates how to set the alua attribute for a LUN (with a preferredport setting) on which ALUA is not enabled. ALUA must be enabled for the preferredport attribute to be recognized. # lun set container name alua

Removing an ALUA Setting You can disable the alua attribute for a LUN using the following command. # no lun set container name alua Viewing Preferred Port Settings and Sessions You can display a list of LUNs and their preferredport settings using the following command. # show luns preferred port Container: MyContainer LUN Preferred port ---------------------------------------------ALUAlun 1 bigthin 1 thick1 2 thick2 2 thick3 2 thin1 1 thin2 1 thin3 1 thin4 1 You can display a list of preferredport sessions using the following command. # show luns preferredport sessions

342


535-0008-00 Rev 10


Container: MyContainer LUN Preferred port ---------------------------------------------VMEM_1 1 Connected Sessions : vmem-1-mg-a : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 VMEM_1 1 Connected Sessions : vmem-1-mg-b : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 VMEM_2 2 Connected Sessions : vmem-2-mg-a : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2 VMEM_2 2 Connected Sessions : vmem-2-mg-b : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2 Viewing ALUA Settings You can display a list of LUNs with alua settings using the following command. # show luns alua Container: MyContainer LUN ALUA -----------------------------------ALUAlun 1 bigthin 1 thick1 0 thick2 0 thick3 0 thin1 1 thin2 1 thin3 1 thin4 1

535-0008-00 Rev 10


343


You can display a list of alua sessions using the following command. # show luns alua sessions Container: MyContainer LUN ALUA -----------------------------------VMEM_1 1 Connected Sessions : vmem-1-mg-a : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 VMEM_1 1 Connected Sessions : vmem-1-mg-b : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 1 VMEM_2 1 Connected Sessions : vmem-2-mg-a : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2 VMEM_2 1 Connected Sessions : vmem-2-mg-b : Port: hba-a1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2 : Port: hba-b1, Session: wwn.21:00:00:24:ff:45:fc:9a, LUN ID: 2

Thin Provision LUNs You can create a thin provision LUN within a specified storage container by adding the thin option to the end of the lun create command.


a substantial performance hit. For more information, see Optimizing Performance with ALUA on page 341.

Creating a Thin Provision LUN The command for creating a thin provision LUN provides the same options as the command for creating a regular LUN, with the exception of thin at the end of the syntax. # lun create container name size [ | equal] [quantity ] [nozero] [readonly] [startnum ] [blocksize 512 | 4096] [offline] thin

344


535-0008-00 Rev 10


The following example demonstrates the creation of a 10 GB thin provision LUN named Mylun1 in a container named “apple”. # lun create container MyContainer name Mylun1 size 10 thin LUN create: success!

Modifying Read/Write Access for a Thin Provision LUN The lun set command allows you to specify a LUN as read-only. This command is the same as that used to modify the read/write access for a regular (thick) LUN. # lun set container name readonly

Renaming a Thin Provision LUN You can use the lun rename command to rename a regular (thick) LUN. # lun rename container name to

Exporting a Thin Provision LUN The process for exporting thin provision LUNs is the same as exporting regular (thick) LUNs. For more information, see Exporting LUNs on page 193.

Note: Once a LUN is exported, it cannot be deleted without first unexporting the LUN.

# lun export container name [lunid][to ]... [using ]

Unexporting a Thin Provision LUN The process for unexporting thin provision LUNs is the same as exporting regular (thick) LUNs. For more information, see Exporting LUNs on page 193.


LUN.

# no lun export container name

Showing a List of Thin Provision LUNs You can view information about thin provision and thick LUNs using the show luns command. You can filter LUNs by container, name, and session parameters. To view a list of containers, enter show luns container ?. The name parameter enables you to view only those LUNs which are prefixed by a particular name.

535-0008-00 Rev 10


345


# show luns container name sessions The following example shows the results for thin and thick LUNs. # show luns Container: MyContainer LUN Size Type RW Bksz Status Ports --------------------------------------------------------------------Mylun1 10G thin rw 512 HA A,B Mylun2 5G thick rw 512 HA A,B

Deleting a Thin Provision LUN This delete LUN command is the same for thin and thick (regular) LUNs.


LUN.

# no lun create container name

Snapshots A snapshot is a space-efficient, point-in-time copy of a LUN or other storage volume. By default, snapshots are read-only. However, you can set the permissions on snapshots to read-write.

Note: An application has to be quiesced (on the client side) before you can take an application consistent snapshot. The snapshots discussed in this section refer to crash consistent snapshots, taken on the array.

Creating Snapshots A snapshot name must start with alphanumeric characters, and may also include dashes and underscores. The [description] option allows you to add a description to the snapshot. You can allow write privileges to the snapshot with the [readwrite] option. The [protect] option prevents the snapshot from being deleted during space reclamation. # snapshot create container lun name [description][readwrite][protect] Deleting Snapshots The following rules apply when deleting snapshots:

346

•

Snapshots that belong to a group cannot be deleted individually. However, you can delete all the snapshots in a group.

•

Snapshots that have been exported cannot be deleted. You must unexport the snapshot first, then it can be deleted. Violin 6000 Series Memory Array User’s Guide

535-0008-00 Rev 10


•


# no snapshot create container lun name

Showing a List of Snapshots Use the following command to display a list of snapshots on the array. Along with the list of snapshots, the name, timestamp, read-only/read-write status, and protected information for each snapshot is displayed. # show snapshots

Container: 41238F00608 Snapshot LUN RW Protect Status Ports AllocSz Created --------------------------------------------------------------------------------------snapg1_lun1_20130416_133533 lun1 ro no HA A,B 0G 2013/04/16_13:35:33 snapg1_lun2_20130416_133533 lun2 ro no HA A,B 0G 2013/04/16_13:35:33 snapg1_lun3_20130416_133533 lun3 ro no HA A,B 0G 2013/04/16_13:35:33

Showing a List of Snapshots for a LUN Use the following command to display a list of snapshots for a specific LUN. # show snapshots container lun

Showing a List of Snapshot Sessions Use the following command to display a list of snapshot sessions: # show snapshots sessions

Showing Snapshots with Associated ALUAs Use the following command to display a list of snapshots and their associated ALUAs. For information on ALUA and preferred ports, see Optimizing Performance with ALUA on page 341. # show snapshots alua

535-0008-00 Rev 10


347


Container: MyContainer snapshot ALUA ------------------------------vmem_20130319_175400 0 lun1_20130319_175700 1 lun2_20130319_180000 1 lun3_20130319_180301 0 lun4_20130319_180601 0 HR_1_20130319_181201 0 Eng1_20130319_181500 1

Showing Snapshots with Associated NACAs Use the following command to display a list of snapshots and their associated NACAs. For more information on NACA, see AIX Compatibility NACA Bit Support on page 179. # show snapshots naca Container: MyContainer snapshot NACA ------------------------------Admin_20130329_5500 0 lun5_20130329_5700 1 lun7_20130329_5800 1 lun3_20130329_5301 0 lun9_20130329_5601 1 HR_1_20130329_5201 0 Eng1_20130329_5900 1

Showing a List of Exported Snapshots Use the following command to display a list of exported snapshots. # show exports

Container: MyContainer LUN Snapshot Target Initiator LUNID Status --------------------------------------------------------------------Admin1_thick 20813 iscsi iqn.1994-05.com:01f775debf9 1 Active lun05_thin 20814 iscsi iqn.1994-05.com:01f775debf9 3 Active lun09_thin 20815 iscsi iqn.1994-05.com:01f775debf9 4 Inactive HR13_thick 20816 iscsi iqn.1994-05.com:01f775debf9 2 Active

348


535-0008-00 Rev 10


Performing a Rollback Before you try performing a rollback, you should understand the requirements and the process. Requirements: a. If the snapshot you want to restore the LUN to is read-write, the snapshot must be taken

offline first. b. After the snapshot is taken offline, the LUN you want to restore must be taken offline. Process:

A rollback follows this process: a. When a rollback is initiated, the system creates a new snapshot of the LUN to preserve

the state of the LUN prior to the rollback. b. The LUN is changed to match the contents of the (restore) snapshot specified in the rollback. c. The LUN can be brought back online, as well as the (restore) snapshot used in the rollback process. Note: The system snapshot—taken at the beginning of the rollback process—can later be used

to undo the rollback, should the need arise.

Command Syntax

# snapshot rollback set container lun name Renaming Snapshots You can change the name of a snapshot at any time, provided it has not been exported. Changing the name of a snapshot does not impact the existing sessions for that snapshot, as long as the snapshot has not yet been exported. Specify the container ID, LUN name, old snapshot name, and new snapshot name. # snapshot rename container lun name to Modifying the Access to Snapshots You can change snapshot access from read-only to read-write, or from read-write to read-only.

Note: Existing snapshot sessions are disconnected after making this change.

You are not allowed to modify the individual properties of a snapshot that belong to a snap group. For example, you are not allowed to change the read-only/read-write property or the protect property of a snapshot that belongs to a snap group.

535-0008-00 Rev 10


349


# snapshot set container lun name readwrite

Basic Syntax for Exporting a Snapshot # snapshot export container lun [snapshotname] [lunid] [to ]... [using ]

Assigning a LUN ID to an Exported Snapshot # snapshot export container lun [snapshotname] [lunid]

Exporting a Snapshot to an Initiator or Initiator Group # snapshot export container lun [snapshotname] [to ]

Exporting a Snapshot Using a Target Port # snapshot export container lun [snapshotname] [using ]

Unexporting a Snapshot # no snapshot export container lun [snapshotname]

Protecting a Snapshot # snapshot set container lun name protect

Unprotecting a Snapshot # no snapshot set container lun name protect

Snap Groups You can create a snap group from an established LUN group. A snap group is a group of snapshots of the LUNs belonging to the LUN group. Creating a Snap Group This operation creates a snapshot of each LUN in the snap group, then links the snapshots together into a group. # snapgroup create container lungroup name [readwrite] [protect] [description]

350


535-0008-00 Rev 10


Deleting a Snap Group You can delete a snap group using its name. When you delete a snap group, all the snapshots contained within the group are deleted.

Note: Snapshots that belong to a group cannot be deleted individually. However, you can delete

the entire group of snapshots. # no snapgroup create container lungroup name

Viewing Snap Group Information If you use the show snapgroups command without any options, the resulting list displays all snap groups for all LUN groups on all containers. # show snapgroups [container] [lungroup-name] [snapgroup-name] [protected] [readwrite]

# show snapgroups container 41238F00608 lungroup lung1 snapgroup snapg0 Container: 41238F00608 Lungroup: lung1 Description: Just another snapgroup # of snapshots: 3 Snapshots ---------------------------------------------------------------------snapg1_lun1_20130416_133533 snapg1_lun2_20130416_133533 snapg1_lun3_20130416_133533 Total Allocated Size: 0G Snapshot Protected/Unprotected: 0G/0G Rolling Back the Contents of a Snap Group You can roll back (restore) the contents of all LUNs in a LUN group from a snap group. This operation does not require that the LUNs be unexported. However, all the LUNs and snapshots involved are taken offline when the contents of the LUN are changed. The following command specifies the LUN group to be restored by the specified snap group. # snapgroup rollback container lungroup name

535-0008-00 Rev 10


351


Adding or Changing a Snap Group Description Use the following command to add a description to a snap group, or change an existing description. Enter the new description text in place of the variable. # snapgroup set container lungroup name description

Changing a Snap Group Port Use the following command to change the port to which the snap group is exported. # snapgroup set container lungroup name port

Protecting and Unprotecting a Snap Group Use the following command to protect a snap group from being automatically deleted when free space is needed. # snapgroup set container lungroup name protect Use the following command to allow the snap group to be automatically deleted when free space is needed. # no snapgroup set container lungroup name protect

Changing Snap Group Permissions Use the following command to grant read/write access to a snap group. # snapgroup set container lungroup name readwrite Use the following command to set the permissions to read-only for a snap group with read/write access. # no snapgroup set container lungroup name readwrite

352


535-0008-00 Rev 10


Renaming a Snap Group Use the following command to change the name of an existing snap group. You must specify the LUN group associated with the snap group. # snapgroup rename container lungroup name to

Exporting a Snap Group to an Initiator Use the following command to export a snap group to a specified initiator. # snapgroup export container lungroup name [snap_group_name] to

Exporting a Snap Group Using a Port Use the following command to export a snap group using a specified port or ports. # snapgroup export container lungroup name [snap_group_name] using

Unexporting a Snap Group Use the following command to unexport a snap group. # no snapgroup export container lungroup name [snap_group_name]

Schedule Management You can automatically schedule the creation of snapshots and snap groups using schedule management commands. Creating a Schedule Use the following command syntax to create a schedule: # schedule snapshot create container Use the following to customize a schedule: — — — — — — —

535-0008-00 Rev 10

description: Adds an identifying description to the schedule disable: Specifies a schedule as disabled until the schedule is enabled end-date: Specifies the end date for the schedule end-time: Specifies the end time for the schedule lun: Creates a schedule for a LUN lungroup: Creates a schedule for a LUN group max-keep: Specifies the number of snapshots kept before deleting; does not apply for a one-time schedule


353


— protect: Specifies the snapshots taken by this schedule are protected from automatic deletion — readwrite: Specifies that the snapshots taken by this schedule have read/write access — schedule-name: Specifies a name for the schedule — start-date: Specifies the start date for the schedule — start-time: Specifies the start time for the schedule — time-of-day: Specifies the time of day at which a non-periodic schedule triggers — type-daily: Specifies a daily schedule [time-of-day] — type-monthly: Specifies a monthly schedule [day-of-month] [interval] [day-of-month]: Specifies the day of the month on which a monthly schedule executes; use a negative number to count backwards from the end of the month [interval]: Specifies the number of months for the schedule to repeat; the default is 1 if this option is not specified — type-once: Specifies a one-time schedule [date] [date]: Specifies the execution date for a one-time schedule — type-periodic: Specifies a periodic schedule [periodicity] [periodicity]: Specifies the number of recurring intervals for the schedule — type-weekly: Specifies a weekly schedule [days-of-week] [days-of-week]: Specifies the days of the week on which a weekly schedule executes Deleting a Schedule Deleting a schedule does not affect (or impact in any way) the snapshots or snap groups already created by the schedule. If a schedule is currently in the process of creating a snapshot or snap group, it does not impact the creation. # no schedule snapshot create container

Basic Syntax for Modifying a Schedule Use the following command with the necessary options to modify an existing schedule. # schedule snapshot modify container lun|lungroup

Changing a Schedule Name The following example changes the name of a schedule associated with the specified LUN. # schedule snapshot modify container lun schedule-name new-schedule-name

Changing the Maximum Number of Kept Snapshots The following example changes the maximum number of snapshots that are kept. After this number is reached, the next oldest snapshot is deleted when a new snapshot is created. # schedule snapshot modify container lun schedule-name max-keep

354


535-0008-00 Rev 10


Enabling a Schedule The following example enables a schedule. # schedule snapshot modify container lun schedule-name enable

Disabling a Schedule The following example disables a schedule. # schedule snapshot modify container lun schedule-name disable Note: You can disable—temporarily pause—a schedule at any time. If a schedule is in progress,


Making a Schedule Read-Only The following example sets the permissions for the snapshots created by the schedule as read-only. # schedule snapshot modify container lun schedule-name readonly

Making a Schedule Read-Write The following example sets the permissions for the snapshots created by the schedule as readwrite. # schedule snapshot modify container lun schedule-name readwrite

Protecting Snapshots in a Schedule The following example protects the snapshots created by the schedule so they are not automatically deleted by the system to create more free space. # schedule snapshot modify container lun schedule-name protect

Unprotecting Snapshots in a Schedule The following example allows the snapshots created by the schedule to be automatically deleted by the system to create more free space. # schedule snapshot modify container lun schedule-name unprotect

535-0008-00 Rev 10


355


Changing the Time of Day a Schedule Executes The following example changes the time of day a schedule executes. # schedule snapshot modify container lun schedule-name time-of-day

Viewing Schedules Use the following command to view a particular schedule. # show schedule If a snapshot is created by a schedule, it appears in the list of objects. In this case, the schedule name appears in the snapshot list.

Free Space Management You can monitor and manage the available space on containers for LUNs, snapshots, snap groups, and LUN groups.

Viewing a Container Space Summary You can view a summary of the total space and free space on a container, along with the number of LUNs, and the provisioned space in Gigabytes. Ports, host, and status information is also displayed, using the following command. You must be in enable mode to run this command. # show containers Container LUNs Total Free Provisioned Ports HostA HostB Status ----------------------------------------------------------------41238F 8 20685G 20661G 40G A,B v-mga v-mgb HA

Viewing Container Space Utilization You can view a detailed breakdown of space utilization on a container with the show containers space_utilization command. The following terms apply:

•

provisioned size: amount of space asked for

•


•

free space: amount of space available for use

You must enter enable mode to run this command.

356


535-0008-00 Rev 10


# show containers space-utilization name: MyContainer Category Allocated Provisioned ------------------------------------------------Thick LUNs 518G 518G Thin LUNs 3G 185G Snapshots 5G 100G Reserved 154G Container 680G 803G Maximum space 10365G 31097G

Viewing LUN Space Utilization You can view space utilization data for individual LUNs as well as all LUNs on a container. This command displays detailed information about all the LUNs on a container. The following terms apply:

•


•

total (provisioned) size: amount of space asked for

You must enter enable mode to run these commands. # show luns Container: MyContainer LUN Size Type RW Bksz Status Ports Snapshots AllocSz TotalSz ---------------------------------------------------------------------lun5_thck 50G thick rw 512 HA A,B 324 50G 50G lun6_thck 75G thick rw 512 HA A,B 25 75G 75G lun7_thck 60G thick rw 512 HA A,B 202 60G 60G lun23_thn 35G thin rw 512 HA A,B 121 21G 23G lun24_thn 45G thin rw 512 HA A,B 221 12G 15G lun25_thn 50G thin rw 512 HA A,B 355 22G 27G The following command displays information about a particular LUN on a specified container. # show luns container name Container: MyContainer LUN Size Type RW Bksz Status Ports Snapshots AllocSz TotalSz --------------------------------------------------------------------lun3_thn 50G thin rw 512 HA A,B 554 32G 38G Container: MyContainer Lun: lun3_thn Total Allocated Size: 32G Lun Allocated/Provisioned Size: 32G/50G Lun Snapshot Protected/Unprotected/Shared Size: 2G/4G/6G

535-0008-00 Rev 10


357


Viewing Snapshot Space Utilization You can view space utilization data for snapshots on a container. The following term applies:

•


You must enter enable mode to run this command. # show snapshots Container: MyContainer Snapshot LUN RW Protect Status Port AllocSz ---------------------------------------------------------------------lun20_thck_130324_140 lun20_thck ro no HA A,B 50G lun24_thck_130324_151 lun24_thck ro no HA A,B 80G lun27_thck_130124_157 lun27_thck ro yes HA A,B 40G lun26_thn_130324_161 lun26_thn ro no HA A,B 22G lun28_thn_130324_164 lun28_thn ro no HA A,B 17G lun29_thn_130324_167 lun29_thn ro yes HA A,B 31G

Setting a Used Space Threshold Trigger The following command specifies a percentage of used space in the container that when reached, triggers automatic space reclamation. # container set name usedspace-threshold [soft] [hard]

• •

soft: threshold that issues a warning when broken hard: threshold that starts reclamation when broken

Setting a Provision Threshold Trigger The following command specifies a percentage of provisioned space in the container that when reached, triggers automatic space reclamation. # container set name provision-threshold [soft] [hard]

• •

358

soft: threshold that issues a warning when broken hard: threshold that starts reclamation when broken


535-0008-00 Rev 10

Diagnostic and Display Commands

Diagnostic and Display Commands varray Display array information, including number of VIMMs, temperature, fan speed and LED states. vdiag Display array diagnostic information (must be run from the Master ACM). vincident Display incident report for the array. vinfo Display information about the array. vstat Display array statistics. vvimms Display VIMM information.

Standard Show Commands show alarms Display alarms. See Alarm Reference on page 369 for information about alarms. show array See Array Configuration and Show Commands on page 320. show bootvar Display installed system images and boot parameters. show chassis info Display the following information about the chassis: date, model, serial number and software version. show clock Display the current system time, date, and time zone. See NTP, Clock and Time Zones on page 300 for additional time-specific commands. show email

535-0008-00 Rev 10


359


Display notification settings. This does not include SNMP traps, which are under the snmp-server command tree. See Event Notification on page 302 for additional e-mail-specific commands. show hosts Show all values configured for the following: hostname, name servers, domain name list, and static host mappings. show interfaces Display detailed running state for all interfaces. show inventory Display basic information about the VCMs and VIMMs in the Array. show ip Display IP-related information for the commands above. show log Display logging configuration for the above commands. Where:

Shows current event log file in a scrollable pager.

Shows new event log messages as they arrive.

Shows archived log files.

Shows event logs that match a given regular expression.

Shows event logs that do not meet certain criteria.

show ntp | configuration Display NTP runtime state; display NTP configuration. See NTP, Clock and Time Zones on page 300 for additional NTP-specific commands. show pcie Use the show pcie command to display detailed PCIe connection information. Below is an example of the command and output for a system named violin with four VCMs. The output shows the two possible configurations: direct-attach and mg. See the Violin 6000 Series Memory Array Installation Guide for initial PCIe configuration commands. The first section, called Switch configuration, displays the current routing mode, which in this case is set to the “direct-attach” mode. The default mode (“direct-attach”) is four PCIe direct connections, with two on each ACM. The four PCIe ports run at gen2 speed with a PCIe lane width of x8 lanes per port, as shown in the Port configuration section. The Port configuration section is broken into two parts: one for the ACM in slot “a” and the other for the ACM in slot “b”. The PCIe ports are

360


535-0008-00 Rev 10

Standard CLI Commands

labeled external “a” and “b” below. The display will show optimal for a port that is configured at its highest speed, degraded if it is at a lower speed and no link if it can’t detect a connection. violin-acma [violin: master] (config) # show pcie Current PCI-E config: mg Config mg Config direct-attach Ports acm-a (lab-stein12-acma): mg_out: optimal hba-a: degraded: no link hba-b: degraded: no link mg_in: optimal vcm-a: optimal vcm-b: optimal vcm-c: optimal vcm-d: optimal Ports acm-b (lab-stein12-acmb): mg_out: optimal hba-c: degraded: no link hba-d: degraded: no link mg_in: optimal vcm-a: optimal vcm-b: optimal vcm-c: optimal vcm-d: optimal violin-acma [violin: master] (config) #█ show version Display version information for the current system image. show vimms Display information about the VIMMs. See VIMM Commands on page 324 for additional VIMM-specific commands.

Standard CLI Commands This section describes commands in the following categories:

• • • • • •

Authentication Method and Order on page 362 SSH Configuration on page 362 Banner on page 364 CLI Options on page 364 SNMP Configuration on page 365 Web Interface Configuration on page 366

535-0008-00 Rev 10


361


Authentication Method and Order aaa authentication login default ... no aaa authentication login Sets the list of acceptable authentication methods for system logins. local, radius, and tacacs+ are acceptable methods. The order in which the methods are specified is the order in which the authentication is attempted. aaa authorization map default-user no aaa authorization map default-user When a user is authenticated (via RADIUS or TACACS+) and does not have a local account, this command specifies what local account the authenticated user will be logged on as. If the user name is local, this mapping is ignored. This mapping is used depending on the setting of the “authorization map order”. aaa authorization map order {remote-only,remote-first,local-only} no aaa authorization map order Can be set to one of three choices: “remote-first”, “remote-only”, or “local-only”. Used when authenticating users via RADIUS or TACACS+. The order determines how the remote user mapping behaves. If the authenticated user name is valid locally, no mapping is performed. The setting has the following three possible behaviors:

•

remote-first: If a local-user mapping attribute is returned and is a valid local user name, map the authenticated user to the local user specified in the attribute. Otherwise, if the attribute is not present or not valid locally, use the user specified by the default-user command. (This is the default behavior.)

•

remote-only: Only try to map a remote authenticated user if the authentication server sends a local-user mapping attribute. If the attribute does not specify a valid local user, no further mapping is tried.

•

local-only: All remote users will be mapped to the user specified by the aaa authorization map default-user command. Any vendor attributes received by an authentication server are ignored.

show aaa Shows the current authentication and authorization settings.

SSH Configuration SSH Server [no] ssh server enable Enable or disable the ssh server. If the ssh server is disabled, the CLI is only accessible over the serial console. Note that this does not terminate existing ssh sessions; it will only prevent new ones from being established. ssh server host-key generate

362


535-0008-00 Rev 10


Regenerate new host keys for the SSH server. This generates three keys: RSA for sshv1, RSA for sshv2, and DSA for sshv2. Note that the system automatically generates the host keys on its first boot, so this only needs to be done if a security breach is suspected and the keys need to be changed. ssh server host-key private-key ssh server host-key public-key Manually set the host-key (either private or public, but should be both if changing) of the specified key type. [no] ssh server listen enable Enable (or disable) the listen interface restricted list for sshd. If enabled and at least one non-DHCP interface is specified in the list, the SSH connections are only accepted on those specified interfaces. When disabled, SSH connections are accepted on any interface. [no] ssh server listen interface Provides a mechanism to add and remove interfaces to the 'listen' list. If the interface is also running as a DHCP client, it will be as if the interface was not added. If DHCP is later turned off on this interface, it will be as if the interface was then added to the listen list. show ssh server Display information about the ssh server, including whether or not it is enabled, and the host keys. SSH Client ssh client user identity generate [passphrase ] Generate a new identity (private and public keys) for the specified user name. The given user name must correspond to a valid local user account. When the keys are generated, the private key is written to the user’s ssh directory in an appropriately named file (id_dsa). This identity can be used when the user uses the slogin command to connect from the system to another host. DSA and RSA v2 keys for SSHv2 can be generated. This is specified with dsa2 or rsa2 as the key-type parameter. ssh client user identity public-key ssh client user identity private-key Set the public or private key (of specified type) for the specified user name. This is an alternative to generating the key in the above command and is also used for reverse mapping generated keys. no ssh client user identity [] Removes the public/private keys for the specified user. Any private key file in a valid user SSH directory is deleted. ssh client user authorized-key sshv2

535-0008-00 Rev 10


363


The specified key is added to the list of authorized sshv2 RSA or DSA public keys for this user account. These keys can be used to log into the user’s account. The specified user must be a valid account on the system. As keys are added, an implicit id is associated with the key. This is to make key deletion easier. Be aware that if a key is being pasted from a cut buffer and was displayed with a paging program, it is likely that newline characters have been inserted, even if the output was not long enough to require paging. Most likely show command output will be displayed this way, as paging is enabled by default in the CLI. One can specify “no cli session paging enable” before doing the show command to prevent the newlines from being inserted. no ssh client user authorized-key sshv2 Remove a public key from the specified user’s authorized key list. The key identifier can be found by using 'show ssh client'. show ssh client Display information about SSH client identities (public/private keys) and the per user list of authorized keys for the users.

Banner At various login points, some legal and welcome text can be displayed. This text is controlled by bindings and can be set by the user from the CLI as follows: banner login no banner login Set the contents of the /etc/issue and /etc/issue.net files. banner motd no banner motd Set the contents of the /etc/motd file. show banner Display the contents of the currently configured banners.

CLI Options There are four groups of commands relating to the CLI itself:

364

•

cli session ... commands change a setting only for the current CLI session. They do not affect any other sessions, and can be performed by any user at any time. All of these commands are INTERACTIVE (not available from the Scheduler).

•

cli default ... commands change the defaults for the specified setting for all future CLI sessions of all users. They also change the setting for the current session from which they were executed, but not for any other currently active sessions. Since they change configuration, the user must be in Config mode to run them, and hence they can only be run by admin.


535-0008-00 Rev 10


•

Other cli ... commands which take one-time actions, rather than change a setting, and thus do not fall under the session or default umbrellas. For example, “cli clear-history”.

•

terminal ... commands are clones of a subset of the cli session ... commands, and are only present for Cisco compatibility. All of these commands are INTERACTIVE (not available from the Scheduler).

Note that some settings, such as the terminal length and width, are inherently session-specific, and there are no corresponding commands to set the defaults in configuration. cli default auto-logout cli session auto-logout no cli default auto-logout no cli session auto-logout Control the length of user inactivity required before the CLI will automatically log a user out. The no ... variants of this command disable the automatic logout feature.

SNMP Configuration SNMP MIBs are available at ftp://ftp.violin-memory.com and in the Violin Web interface. [no] snmp-server enable Enable or disable the SNMP server. Note that this not only stops serving of SNMP variables, but also the sending of SNMP traps. [no] snmp-server enable traps Enable or disable sending of SNMP traps from this system. Traps may only be enabled if the SNMP server overall is enabled. The traps sent by the SNMP agent are:

• • • • • • •

Cold boot (may include SNMP configuration having been changed) Link up/down CPU load too high CPU load no longer too high Paging activity too high A process has failed A process has exited unexpectedly

Note that traps are only sent if there are trap sinks configured with the snmp-server host ... command, and if these trap sinks are themselves enabled. [no] snmp-server listen enable Enable (or disable) the listen interface restricted list for snmpd. If enabled and at least one nonDHCP interface is specified in the list, the snmp connections are only accepted on those specified interfaces. When disabled, snmp connections are accepted on any interface. [no] snmp-server listen interface

535-0008-00 Rev 10


365


Provides a mechanism to add and remove interfaces to the “listen” list. If the interface is also running as a DHCP client, it will be as if the interface was not added. If DHCP is later turned off on this interface, it will be as if the interface was then added to the listen list. snmp-server community no snmp-server community Set the community name required to be supplied with SNMP requests to the system. snmp-server contact no snmp-server contact snmp-server location no snmp-server location Set the syscontact and syslocation variable served from the System MIB in MIB-II. snmp-server host traps [version ] no snmp-server host Add or remove a host to which SNMP traps should be sent. Note that this setting is only meaningful if traps are enabled, though the list of hosts may still be edited if traps are disabled. [no] snmp-server host disable Disable a trap sink without actually removing it altogether from the configuration. All trap sinks are created enabled. [no] snmp-server traps event Specify which types of events should be sent as SNMP traps. By default the entire list of notifiable events are sent as SNMP traps to any declared trap sinks. This command enables or disables a single event for conversion to an SNMP trap. show snmp Display all SNMP configuration options.

Web Interface Configuration [no] web enable Enable or disable the Web Interface. [no] web http enable Enable or disable HTTP access to the Web Interface. This setting is only meaningful if the Web Interface as a whole is enabled. web http port no web http port Set the port number for HTTP. The default is 80. The no command resets it to the default, but does not disable HTTP.

366


535-0008-00 Rev 10


[no] web https enable Enable or disable HTTPS (HTTP over SSL) access to the Web Interface his setting is only meaningful if the Web Interface as a whole is enabled. web https port no web https port Set the port number for HTTPS. The default is 443. The no command resets it to the default, but does not disable HTTPS. [no] web httpd listen enable Enable (or disable) the listen interface restricted list for HTTPD. If enabled and at least one nonDHCP interface is specified in the list, the http connections are only accepted on those specified interfaces. When disabled, http connections are accepted on any interface. [no] web httpd listen interface Provides a mechanism to add and remove interfaces to the “listen” list. If the interface is also running as a DHCP client, it will be as if the interface was not added. If DHCP is later turned off on this interface, it will be as if the interface was then added to the listen list. web auto-logout no web auto-logout Control the length of user inactivity required before the Web Interface will automatically log out a user. “no web auto-logout” disables the automatic logout feature. web session renewal no web session renewal Control the length of time before Web session cookies are automatically regenerated. web session timeout no web session timeout Control the maximum lifetime of a Web session cookie. show web Display Web Interface configuration settings.

535-0008-00 Rev 10


367


368


535-0008-00 Rev 10

Alarm Reference

APPENDIX B

This chapter describes how you may use Memory Array tools to diagnose system and VIMM-specific alarms. The chapter describes non-service affecting and service affecting alarms that may occur during operation of the Violin Memory Array and provides information on resolving those alarms in three sections.

• • •

Understanding Memory Array Alarms on page 369 Service-Affecting Alarms on page 372 Non-Service Affecting Alarms on page 377

Understanding Memory Array Alarms The Memory Array provides you with key tools for identifying and interpreting raised alarms.

•

Alarms are indicated by the state of the LEDs on the front of the Violin 6000 Series Memory Array chassis.

•

Information about an active alarm may then be viewed by running the show alarms command.

•

Alarms for Violin 6000 Series Memory Array modules are also shown on the Array Status screen in the Violin Web Interface. See Showing Status of the Memory Array Modules on page 149 for more information.

This appendix provides information about how to resolve and clear these alarms when they occur.

535-0008-00 Rev 10


369

Appendix B - Alarm Reference

Interpreting Alarm LEDs The current status of the Violin 6000 Series Memory Array system may be indicated when a service LED enters an alarmed state. Faulty modules can be replaced with no downtime. Six LEDs— identified below—are located on the front of the Violin 6000 Series Memory Array chassis.

AC-A

AC-B

Interface Status ID LED Button

System Status

VIMM Status

Figure B.1 Front Status LEDs

The System Status and VIMM Status LEDs alert you of System and VIMM-specific alarms by their color and state: a solid amber light. When both LEDs are solid amber, an alarm has been registered for one or more VIMMs. A solid amber System Status LED by itself indicates that an alarm has been registered for an issue unrelated to a VIMM. See System LEDs on page 15 for information about the LEDs not shown below.

LED

System Status

VIMM Status

Interface Status

LED State

Description

Green

The following system components are operating normally: VIMM, internal Memory Gateway, HBA, vRAID Controller Module, Array Controller Module, Fan, Front Panel Module, Power Controller Module.

Amber

An alarm has been raised on at least one of the following system components: VIMM, internal Memory Gateway, HBA, vRAID Controller Module, Array Controller Module, Fan, Front Panel Module, Power Controller Module.

Grey

The system is off or starting up.

Green

All VIMMs are functioning correctly.

Amber

An alarm has been raised for at least one VIMM. A VIMM is not powered or not present.

Green

All interfaces are OK.

Table B.1 Front LED Descriptions

Many alarms occur during bootup and automatically clear as soon as the system stabilizes. For example, the Data Plane Unavailable alarm frequently occurs during bootup because the data plane is not enabled until later in the process. Such alarms may be safely ignored.

370


535-0008-00 Rev 10

Understanding Memory Array Alarms

Understanding System Alarm Commands When hardware malfunctions or another error occurs on the Violin 6000 Series Memory Array system, you are notified in the following ways: the System and/or VIMM Status LED on the front of the Memory Array chassis is lit solidly and an alarm is raised by the system. The names of active system alarms may be viewed by running the show alarms command on the Memory Array. show alarms command The show alarms command displays the current alarms on a Memory Array. The command may be run by entering show alarms at the root context level. violin-acmb-acm-b [violin: master] (config) # show alarms --- show alarm for violin-acmb at Thu May 10 15:25:03 2012 No ACM alarms. VCM alarms: vcm-a: No alarms vcm-b: No alarms vcm-c: No alarms vcm-d: No alarms No VIMM alarms. No MG alarms. No HBA alarms. No FAN alarms. violin-acmb-acm-b [violin: master] (config) # The show alarms command returns the name of every active alarm on the Violin 6000 Series Memory Array system. Instructions for clearing system alarms are offered in this appendix.

535-0008-00 Rev 10


371


Service-Affecting Alarms Service-affecting alarms are alarms raised by the system when an error occurs that interrupts traffic and indicates that the data plane is either severely impacted or completely disabled. This section is organized into eight topics:

• • • • • • • •

Equipment Service-Affecting Alarms on page 372 Flash Service-Affecting Alarms on page 372 Port Service-Affecting Alarms on page 373 Power Service-Affecting Alarms on page 373 RAID Service-Affecting Alarms on page 374 System Service-Affecting Alarms on page 375 Temperature Service-Affecting Alarms on page 376 VIMM Service-Affecting Alarms on page 376

Equipment Service-Affecting Alarms There is one service affecting equipment alarm:

•

Booting Suspended, Fans Missing Alarm on page 372

Booting Suspended, Fans Missing Alarm The Booting Suspended, Fans Missing alarm indicates that booting has been suspended because an insufficient number of fans were detected. The alarm indicates the number of fans missing.

Flash Service-Affecting Alarms There are two service affecting flash alarms:

• •

Fatal Flash Error, Alarm on page 372 Unformatted System Alarm on page 373

Fatal Flash Error, Alarm The Fatal Flash Error, alarm represents an error code associated with the failure. The error is representative of a fatal error in the flash datapath. The system must shut down because the non-functional VIMM has made the current VIMM configuration invalid. To clear the alarm, reboot the system. Contact Violin Memory Customer Support for assistance.

Note: Before performing a reboot, stop any activity that may be affected by the interruption, such

as applications currently accessing the system.

372


535-0008-00 Rev 10

Service-Affecting Alarms

Unformatted System Alarm The Unformatted System alarm indicates that all the VIMMs in the system are unformatted. The system is unusable until the system VIMMs are formatted. The system should be formatted if there is no data on the system.

Port Service-Affecting Alarms There are two service affecting port alarms:

• •

No PCIe Host Detected on Enabled Port Alarm on page 373 Port Negotiated to 0 Lanes Alarm on page 373

No PCIe Host Detected on Enabled Port Alarm The No PCIe Host Detected on Enabled Port alarm indicates that one of the ACMs is not present and may be malfunctioning. Port Negotiated to 0 Lanes Alarm The Port Negotiated to 0 Lanes alarm indicates that the PCIe interface between the VCM and ACM negotiated to 0 lanes. The Interface Status LED flashes. To clear the alarm, reboot the system. Contact Violin Memory Customer Support for assistance or if the problem persists.



Power Service-Affecting Alarms There are two service affecting power alarms:

• •

AC Power Failure, Shutting Down Alarm on page 373 Main Power Failure, System Shut Down Alarm on page 373

AC Power Failure, Shutting Down Alarm The AC Power Failure, Shutting Down alarm indicates a failure in the power feeds to the power supplies. Check the AC supply circuits. Main Power Failure, System Shut Down Alarm The Main Power Failure, System Shut Down alarm indicates a failure detected in the main power supplies.

535-0008-00 Rev 10


373


This should only occur in the unlikely event of both power supplies failing. At least one power supply should be replaced. Contact Violin Memory Customer Support for assistance.

RAID Service-Affecting Alarms There are five service affecting RAID alarms:

• • • • •

FATAL RAID Error Alarm on page 374 Multiple RAID Errors Alarm on page 374 RAID Group Faulted Alarm on page 374 RAID Consistency Error Alarm on page 374 RAID Group Faulted Alarm on page 374

FATAL RAID Error Alarm The FATAL RAID Error alarm indicates that a fatal RAID group error has been detected at the displayed vector. A properly configured RAID group configuration is necessary for the system to be operational. To clear the alarm, reboot the system. Contact Violin Memory Customer Support for assistance.



Multiple RAID Errors Alarm The Multiple RAID Errors alarm indicates that more than one RAID error has occurred and at what location. This is a nonrecoverable error. Contact Violin Memory Customer Support for assistance. RAID Consistency Error Alarm The RAID Consistency Error alarm indicates that there is a consistency error with the RAID grouping and at what location. This is a nonrecoverable error. Contact Violin Memory Customer Support for assistance. RAID Group Faulted Alarm The RAID Group Faulted alarm indicates that two or more VIMMs in the specified RAID group have failed and no spares are available to rebuild the RAID group. To clear the alarm, replace the failed VIMMs and insert spares. RAID V2P Error Alarm The RAID V2P Error alarm indicates that the RAID group has experienced an error with a VIMM at the specified ID number. This is a nonrecoverable error. Contact Violin Memory Customer Support for assistance. 374


535-0008-00 Rev 10

Service-Affecting Alarms

System Service-Affecting Alarms There are six service affecting system alarms:

• • • • • •

Data Plane Unavailable Alarm on page 375 I/O Timeout From Host Alarm on page 375 Scheduler Paused Alarm on page 375 Unable to Poll VIMM Temperatures; System Shutdown Alarm on page 375 System Booting Alarm on page 376 System Shut Down Alarm on page 376

Data Plane Unavailable Alarm The Data Plane Unavailable alarm indicates that the data plane is unavailable. This alarm occurs under various conditions, including during a system boot if the system experiences an unrecoverable PCIe error, and if the host does not receive an I/O response within a specific time interval. If the alarm occurs on bootup, this alarm is self-clearing and clears upon successful completion of the bootup process. To clear this alarm if it occurs because of an unrecoverable PCIe error, check your PCIe connections and reboot the system and the host. Contact Violin Memory Customer Support for assistance.



I/O Timeout From Host Alarm The I/O Timeout From Host alarm indicates that an I/O request from the host to the Violin 6000 Series Memory Array timed out before the request was acknowledged. This time out results in a disabling of the data plane and the System Status LED flashes until the system is restarted. This is a nonrecoverable error. Contact Violin Memory Customer Support for assistance. Scheduler Paused Alarm The Scheduler Paused alarm indicates that the hardware scheduler temporarily paused during bootup or when a VIMM is being taken out of service. To clear the alarm, wait for the system to successfully complete the bootup process. Unable to Poll VIMM Temperatures; System Shutdown Alarm The Unable to Poll VIMM Temperatures; System Shutdown alarm indicates that the management channel is not allowing temperatures on the VIMMs to be polled, which should never happen. The system is shut down to prevent overheating.

535-0008-00 Rev 10


375


System Booting Alarm The System Booting alarm indicates that the system is in the bootup process. To clear the alarm, wait for the system to successfully complete the bootup process. System Shut Down Alarm The alarm indicates the system has been shut down.

Temperature Service-Affecting Alarms There is one service-affecting temperature alarm. Controller FPGA Overheated, System Shut Down Alarm The Controller FPGA Overheated, System Shut Down alarm indicates that the main processor of the system experienced temperatures over the acceptable maximum temperature and has shut the system down to prevent damage to the equipment. The alarm specifies the system’s current temperature in Celsius. To clear the alarm, reboot the system, check the fans for proper operation, and check the ambient temperature of the system. Contact Violin Memory Customer Support for assistance.



VIMM Service-Affecting Alarms The fans operate at high speed and an alarm is raised for a VIMM if the temperature reaches 75°C. The VIMM automatically shuts down (taken out of service) upon a third consecutive reading of 85°C. A system log event is generated when VIMM temperature reaches 75°C (indicating overheating). There are four service affecting VIMM alarms:

• • • •

Insufficient VIMMs Alarm on page 376 Main Controller Does Not Support VIMMs Alarm on page 377 Multiple VIMMs Failed; System Shut Down Alarm on page 377 VIMM Flash Format In Progress Alarm on page 377

Insufficient VIMMs Alarm The Insufficient VIMMs alarm indicates that an insufficient number of VIMMs have been detected to construct a usable system.

376


535-0008-00 Rev 10

Non-Service Affecting Alarms

Main Controller Does Not Support VIMMs Alarm The Main Controller Does Not Support VIMMs alarm indicates that the software version running on the vRAID controller with which the VIMM is associated does not support the specified VIMM type: Flash or DRAM. Multiple VIMMs Failed; System Shut Down Alarm The Multiple VIMMs Failed; System Shut Down alarm indicates that multiple VIMMs are not operational and the system must shut down because the non-functional VIMMs have made the current VIMM configuration invalid. To clear the alarm, replace the non-functioning VIMMs. Call Violin Memory Customer Support for replacement parts. After replacing the VIMMs in the system, the VIMM configuration should once again be valid. VIMM Flash Format In Progress Alarm The VIMM Flash Format In Progress alarm indicates that VIMMs are being formatted. The alarm continues to be raised while the format is in progress and is cleared automatically when the format has completed. If any VIMMs fail to format, an alarm is raised on each VIMM. VIMM Faulted; Do Not Use, Replace VIMM Alarm This alarm identifies the faulty VIMM and the reason for the fault. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide.

Non-Service Affecting Alarms Non-service affecting alarms are alarms that call attention to errors that do not affect data traffic on the system. This section describes the non-service affecting alarms that may be raised on the system and provides instructions for resolving those alarms. This section is organized into seven topics:

• • • • • • •

Temperature Non-Service Affecting Alarms on page 378 Equipment Non-Service Affecting Alarms on page 378 Flash Non-Service Affecting Alarms on page 379 Port Non-Service Affecting Alarms on page 380 Power Non-Service Affecting Alarms on page 380 RAID Non-Service Affecting Alarms on page 380 VIMM Non-Service Affecting Alarms on page 381

535-0008-00 Rev 10


377


Temperature Non-Service Affecting Alarms Two non-service affecting temperature alarms may be raised by the system:

• •

Controller FPGA Overheating Alarm on page 378 High Temperature on Board Sensor Alarm on page 378

Controller FPGA Overheating Alarm The Controller FPGA Overheating alarm indicates that the main processor of the system is experiencing temperatures over the acceptable maximum temperature. The alarm specifies the system’s current temperature in Celsius. If the chassis is cooled down to within the acceptable temperature range, the system will not shut down. To clear the alarm, confirm that the fans are operating correctly, there are no airflow restrictions, the lid is in place, and the ambient temperature is within the acceptable range.

Caution: To avoid overheating and the generation of system alarms, the lid must not be off of

the system for more than 5 minutes when the ambient temperature is 35 °C (95 °F) and above.

See “Replacing a Fan” in the Violin 6000 Series Memory Array Service Guide. High Temperature on Board Sensor Alarm The High Temperature on Board Sensor alarm indicates that the specified board sensor on the system is unreadable and that the system cannot retrieve system environment information. To clear the alarm, check fan operation, the fan filter, and the ambient temperature of the system. See “Replacing a Fan” in the Violin 6000 Series Memory Array Service Guide.

Equipment Non-Service Affecting Alarms There are three non-service affecting equipment alarms:

• • •

Board Sensor Unreadable Alarm on page 378 Fan Missing Alarm on page 378 Fan Running Too Slowly Alarm on page 379

Board Sensor Unreadable Alarm The Board Sensor Unreadable alarm indicates that the specified board sensor is unreadable and that the system cannot retrieve system environment information. To clear the alarm, check for equipment failure and contact Violin Memory Customer Support. Fan Missing Alarm The Fan Missing alarm indicates the specified fan is missing. 378


535-0008-00 Rev 10


Fan Running Too Slowly Alarm The Fan Running Too Slowly alarm indicates that the specified fan is running at a much lower speed than it should. The alarm specifies the top (1) or bottom (2) fan row and the fan within that fan row: left, center, or right. To clear the alarm, replace the fan in the system.

Flash Non-Service Affecting Alarms There are five non-service affecting flash alarms:

• • • • •

Flash CPL IRQs on VIMM Alarm on page 379 Flash Format Failed on VIMM Alarm on page 379 Flash Sequence Number Mismatch Alarm on page 379 Preparing Flash on VIMM Alarm on page 379 VIMM Flash Format In Progress Alarm on page 379

Flash CPL IRQs on VIMM Alarm The Flash CPL IRQs on VIMM alarm indicates that an error has been detected by the Flash VIMM Control Plane. This is a fatal error on the VIMM and it must be taken out of service. Flash Format Failed on VIMM Alarm The Flash Format Failed on VIMM alarm indicates that the format operation on the specified VIMM failed. The VIMM will be taken out of service. Flash Sequence Number Mismatch Alarm The Flash Sequence Number Mismatch alarm indicates that the VIMM sequence numbers in a particular RAID group do not match when data is written to those VIMMs. To clear the alarm, reboot the system. Contact Violin Memory Customer Support for assistance.



Preparing Flash on VIMM Alarm The Preparing Flash on VIMM alarm indicates that flash is being prepared on the specified VIMM for use by the system. This alarm clears once the flash preparations have completed. VIMM Flash Format In Progress Alarm The VIMM Flash Format In Progress alarm indicates that the flash VIMMs formatting is in progress. To begin the flash formatting, the system must be rebooted first. The VIMM Status LED flashes. To clear the alarm: 535-0008-00 Rev 10


379


1. Wait for the system to successfully complete the boot up process. 2. Type show alarms to display the percentage of the flash format progress.

Port Non-Service Affecting Alarms There is one non-service affecting port alarms:

•

Port Is Not Running at the Maximum Data Rate Possible Alarm on page 380

Port Is Not Running at the Maximum Data Rate Possible Alarm The Port Is Not Running at the Maximum Data Rate Possible alarm occurs when the negotiated rate of the connection on the specified port is lower than the maximum possible data rate. This error can occur when an internal PCIe link between the VCM and ACM is not connected at optimal bandwidth, which may indicate a potential problem with either the VCM or ACM.

Power Non-Service Affecting Alarms There are three non-service affecting power alarms:

• • •

Power Supply Missing Alarm on page 380 Power Supply Failure Alarm on page 380 Power Supply: Polling Failure Alarm on page 380

Power Supply Missing Alarm The Power Supply Missing alarm indicates that a power supply is missing. Power Supply Failure Alarm The Power Supply Failure alarm indicates that a power supply has failed. See “Replacing a Power Supply” in the Violin 6000 Series Memory Array Service Guide. Power Supply: Polling Failure Alarm The Power Supply: Polling Failure alarm indicates a failure to poll the specified power supply. Contact Violin Memory Customer Support.

RAID Non-Service Affecting Alarms There are four non-service affecting RAID alarms:

• • •

380

Power Supply: Polling Failure Alarm on page 380 RAID Group Unprotected Alarm on page 381 RAID Rebuild in Progress Alarm on page 381


535-0008-00 Rev 10


•

RAID Rebuild of Group Completed with Errors (No Additional Information Available) Alarm on page 381

RAID Group Unprotected Alarm The RAID Group Unprotected alarm indicates that one VIMM in the specified RAID group has failed and that the group is no longer RAID protected, because no spare VIMMs are available to rebuild the RAID group. To clear the alarm, replace the failed VIMM and insert a spare VIMM. For step-by-step instructions, see “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. The VIMM Status LED turns off when the system returns to a normal operational state and the alarm clears. RAID Rebuild in Progress Alarm The RAID Rebuild in Progress alarm indicates that a RAID group rebuild is in progress for the specified RAID and lists the group VIMM ID that is built. To clear the alarm, wait for the system to complete the RAID group rebuild and return to a normal operational state. RAID Rebuild of Group Completed with Errors (No Additional Information Available) Alarm The RAID Rebuild of Group Completed with Errors (No Additional Information Available) alarm indicates that a RAID rebuild of the specified group failed. Raid Rebuild Of Group Completed with Errors Starting at 4K Address Alarm The Raid Rebuild Of Group Completed with Errors Starting at 4K Address alarm indicates that a RAID rebuild of the specified group failed, as well as the starting address.

VIMM Non-Service Affecting Alarms There are 19 non-service affecting VIMM alarms:

• • • • • • • • • • •

Excessive ECC Errors on VIMM Alarm on page 382 Failed to Validate Configuration of VIMMs, Raid Group Alarm on page 382 No Spare Memory Modules Alarm on page 382 Programming VIMM Alarm on page 382 Unsupported VIMM Configuration Alarm on page 382 VIMM Booting Alarm on page 383 VIMM Configuration Out Of Spec Alarm on page 383 VIMM Failed Alarms on page 383 VIMM Extremely Hot; VIMM Shutdown Alarm on page 384 VIMM Failed Alarm on page 384 VIMM Failed: Excessive ECC Errors Alarm on page 384

535-0008-00 Rev 10


381


• • • • • • •

VIMM Hot Alarm on page 384 VIMM Incompatible Formatted Capacity Alarm on page 385 VIMM in Maintenance State Alarm on page 385 VIMM Isolated Alarm on page 385 VIMM Temperature Unreadable, VIMM Shut Down Alarm on page 385 VIMM Unformatted Alarm on page 385 VIMM Threshold Exceeded Alarm on page 385

Excessive ECC Errors on VIMM Alarm The Excessive ECC Errors on VIMM alarm indicates that the system has detected an excessive number of ECC errors on the specified VIMM. To clear the alarm, replace the VIMM. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. Failed to Validate Configuration of VIMMs, Raid Group Alarm The Failed to Validate Configuration of VIMMs, Raid Group alarm indicates that the system failed to validate the current VIMM configuration. The alarm lists the ID of the invalid VIMMs and the RAID group in which the error has occurred. To clear the alarm, remove and replace the failed VIMMs. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. No Spare Memory Modules Alarm The No Spare Memory Modules alarm indicates that the system no longer contains any available VIMMs. To clear the alarm, add spare VIMMs or replace VIMMs that have failed. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. Programming VIMM Alarm The Programming VIMM alarm indicates that the specified VIMM is currently being programmed by the system software. To clear the alarm, wait for the system to complete programming of the VIMM. Unsupported VIMM Configuration Alarm The Unsupported VIMM Configuration alarm indicates that the current VIMM configuration consisting of a specified number of RAID group(s) is invalid and unsupported. To clear the alarm, replace the missing VIMM or failed VIMMs. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. For partially-populated systems, install the VIMMs in specific slots. If not inserted in the designated slots, the system may have difficulty using all of the VIMMs efficiently. See VIMM Slot Locations on page 23 for guidance.

382


535-0008-00 Rev 10


VIMM Booting Alarm The VIMM Booting alarm indicates that the VIMM is in the process of booting and lists the percentage completed in the boot up process. The alarm clears when the VIMM has completely booted up. VIMM Configuration Out Of Spec Alarm The VIMM Configuration Out of Spec alarm indicates that the current VIMM configuration is invalid. The show vimms summary command and the alarm message lists the number of system VIMMs that are missing, failed, in maintenance, or in an admin down state. To clear the alarm, list the failed VIMMs using the show vimm detail command, and then shut down the system to replace the missing VIMMs. VIMM Failed Alarms A VIMM Failed alarm indicates that the VIMM failed for one of the following reasons, and that it has been taken out of service. Failed to communicate over management interface

The alarm indicates a failure to communicate with the VIMM over the management interface.

High temperature threshold exceeded

The alarm indicates that the VIMM was shut down because it became too hot.

Failed to read temperature

The alarm indicates that there was a failure to read the temperature sensor on the VIMM.

Isolated

The alarm indicates that the system has been isolated.

Dead

The alarm indicates that the VIMM is identified as dead.

Excessive ECC errors

The alarm indicates that the amount of ECC errors has been exceeded.

Incompatible memory configuration

The alarm indicates that the memory was not properly configured.

Formatted capacity is incompatible with system

The alarm indicates that the format of memory is not compatible.

Formatted capacity is zero

The alarm indicates the memory has not been formatted.

Failed to set VIMM ID

The alarm indicates that the ID of the identified VIMM could not be set.

Failed to set context

The alarm indicates that the context was not set.

Fatal Flash CPL IRQ received

The alarm indicates a fatal error on the VIMM, requiring VIMM replacement.

Bad block threshold exceeded

The alarm indicates that the number of bad blocks on the VIMM has exceeded the safe operational threshold.

Boot failure (NIOS_ASSERT raised on software scan)

The alarm indicates a boot-time failure of the VIMM.

Boot failure (NIOS_ASSERT raised on software ready)


535-0008-00 Rev 10


383


Boot failure (software failed to become ready)


Boot failure (software failed to complete scanning)


Boot failure (activating software failed)


Boot failure (cpu memory is not initialized)


Boot failure (holding cpu in reset failed)


Boot failure (releasing cpu from reset failed)


Previous fatal error detected

The alarm indicates that the VIMM had been fatally-errored on a previous boot cycle and must be replaced.

Incompatible flash data type

The alarm indicates that the type of flash (MLC, SLC) is not compatible with the other VIMMs in the system.

Cannot determine flash data type

The alarm indicates that a VIMM is of an unknown flash configuration.

VIMM Extremely Hot; VIMM Shutdown Alarm The VIMM Extremely Hot; VIMM Shutdown alarm indicates that the temperature of the specified VIMM is out of the acceptable temperature range and the VIMM automatically shuts down. The alarm specifies the last read temperature of the VIMM in Celsius. To clear the alarm: 1. Check for any airflow restrictions in the front panel. 2. If there are no airflow issues and the environment is within specifications, replace the VIMM.

See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. VIMM Failed Alarm The VIMM Failed alarm indicates that the specified VIMM has failed and is inoperable. To clear the alarm, replace the VIMM. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. VIMM Failed: Excessive ECC Errors Alarm The VIMM Failed: Excessive ECC Errors alarm indicates that a specific VIMM is taken out of service because it experienced excessive ECC errors. VIMM Hot Alarm The VIMM Hot alarm indicates that the specified VIMM is reaching its maximum temperature within the acceptable temperature range. The alarm specifies the VIMM ID and the current temperature in Celsius. The VIMM shuts down if the temperature reaches the maximum acceptable temperature. To clear the alarm: 384


535-0008-00 Rev 10


1. Check for any airflow restrictions in the front panel. 2. If there are no airflow issues and the environment is within specifications, replace the VIMM.

See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. VIMM Incompatible Formatted Capacity Alarm The VIMM Incompatible Formatted Capacity alarm indicates that the specified VIMM has a formatted capacity that is different from the rest of the system, and is thus not being used. VIMM in Maintenance State Alarm The VIMM in Maintenance State alarm indicates that the specified VIMM is in maintenance state. To clear the alarm, Contact Violin Memory Customer Support for assistance. VIMM Isolated Alarm The VIMM Isolated alarm indicates that the specified VIMM is isolated and not reachable by any other VIMM within the tree. This is caused by the failure of other VIMMs, preventing a path from the main controller to the VIMM. To clear the alarm, replace the failed VIMM(s). See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. VIMM Temperature Unreadable, VIMM Shut Down Alarm The VIMM Temperature Unreadable, VIMM Shut Down alarm indicates that the temperature for the specified VIMM cannot be read and the VIMM has been shut down. To clear the alarm, replace the VIMM. See “Replacing a VIMM” in the Violin 6000 Series Memory Array Service Guide. VIMM Unformatted Alarm The VIMM Unformatted alarm indicates that the specified VIMM is unformatted, and is thus not being used. VIMM Threshold Exceeded Alarm The VIMM Threshold Exceeded alarm indicates that a VIMM has been taken out of service due to an exceeded monitored threshold. VIMM Incompatible Formatted Capacity Alarm The VIMM Incompatible Formatted Capacity alarm indicates that a VIMM has been formatted to a capacity that is incompatible with the system. The VIMM ID is displayed in the alarm.

535-0008-00 Rev 10


385


vRAID Controller Module (VCM) Alarms The following are possible VCM alarms that may appear. In all cases, the solution is provided along with the alarm. A VCM is alarmed if the temperature reaches 75°C and automatically completes a shut down/ failover upon a second consecutive reading of 82°C (at 30 second intervals). Heartbeat Timeout; VCM Failed; Replace VCM Alarm See the Violin 6000 Series Memory Array Service Guide for instructions on replacing the VCM. VCM Not Responding; VCM Communications Failed; Reboot VCM Alarm See Powering Off and On Memory Array Modules on page 240. VCM Faulted; Do Not Use, Replace VCM Alarm See the Violin 6000 Series Memory Array Service Guide for instructions on replacing the VCM. VCM Not Visible to One or More Hosts; Reboot Host Alarm See Powering Off and On Memory Array Modules on page 240. Manufacturing Data Not Valid; Replace VCM Alarm See the Violin 6000 Series Memory Array Service Guide for instructions on replacing the VCM. Manufacturing Data Not Available; Replace VCM Alarm See the Violin 6000 Series Memory Array Service Guide for instructions on replacing the VCM. VCM Faulted; Previous Heartbeat Failure Detected; Do Not Use, Replace VCM Alarm See the Violin 6000 Series Memory Array Service Guide for instructions on replacing the VCM. RAID rebuild due to rebalance in progress This alarm indicates that a RAID group is unbalanced. RAID groups may become inefficient over time if VIMMs are replaced and if other administrator-initiated changes are made. If two or more VIMMs in a RAID group share the same root port, the RAID group is considered “unbalanced” and performs less than optimally. See Rebalancing RAID Groups on page 242 for more information.

386


535-0008-00 Rev 10

Violin Drivers

APPENDIX C

This chapter describes the configuration of a direct-attached Violin 6000 Series Memory Array system using a Linux host. Installation of the Violin Drivers is only necessary if you connect the Violin 6000 Series Memory Array directly to a host machine.

Note: See the Violin Array Controller Driver for Windows Installation Guide for instructions on installing and using the driver for Windows. The guide is available from the Violin Memory Customer Support site.

Topics covered include system prerequisites, the installation of the Violin drivers, the installation of the Violin utilities, and the creation of block device partitions.

• • • • • •

Violin Linux Drivers Overview on page 387 Installing Violin Linux Driver Prerequisites on page 388 Installing Violin Linux Drivers on page 389 Loading and Unloading Linux Drivers on page 394 Installing Violin Utilities on page 396 Configuring Partitions and Storage Devices on page 397

Violin Linux Drivers Overview In a direct-attached configuration, the Violin Linux driver enables the host machine to access block storage on a Violin Memory Array. The Violin Linux driver (vtms-linux-driver) must be installed on the host machine if the Memory Array is to be directly attached to a Linux host.

535-0008-00 Rev 10


387

Appendix C - Violin Drivers

Understanding Violin Linux Driver Types Violin Linux drivers may be implemented as one of three different device types: character devices, block devices or SCSI devices. The Memory Array system supports only one device type at a time. The Violin Linux driver accesses the system as the following device type: Block

The default and preferred driver mode. Allows for file system creation, mounting, and optional host buffer cache usage.

In general, a Violin Linux driver provides block device access to the Memory Array enabling the host to create and mount block storage. This chapter describes the steps required to install and configure Violin Linux drivers as block devices only. If you plan to use the Violin Linux driver to enable character or SCSI connections, contact Violin Memory Customer Support.

Installation and Configuration Overview The Violin Linux driver enables a direct-attached connection between a Violin Memory Array and a Linux host machine. During Phase 3 of a Memory Array deployment, you may choose to configure the system for block storage or direct-attached storage. If you choose direct-attached configuration, you must install the appropriate device driver (Linuxor Windows) and the Violin utilities on the host machine. In general, the configuration of a direct-attached Memory Array system using a Linux host is a fivestep process.

•

Step 1: Installing Violin Linux Driver Prerequisites: Ensure that all of the required packages are installed on the Linux host before you install and build the Violin Linux driver.

•

Step 2: Installing Violin Linux Drivers: Install the Violin Linux driver on the host, build the driver, and configure the host to load the Violin Linux driver manually or as a module.

•

Step 3: Loading the Violin Linux drivers: Depending upon where the Linux drivers were installed, they may be loaded manually or automatically as modules when the host boots up.

•

Step 4: Installing Violin Utilities: Violin utility commands enable you to monitor and fine tune the performance of the Memory Array.

•

Step 5: Configuring Partitions and File Systems: If the Violin Linux driver is used as a block device, you may create partitions and file systems.

Installing Violin Linux Driver Prerequisites The first step towards configuring a direct-attached Violin Memory Array system with a Linux host is to ensure that all of the required packages are installed on the Linux host. Three packages must be installed on the Linux host before you can install the Violin Linux driver: the Kernel Development Package, the GCC Compiler, and the Libaio Development Package. The actual name of each package may vary depending on the Linux distribution.

388


535-0008-00 Rev 10

Installing Violin Linux Drivers

Kernel Development Package The Kernel Development Package provides the kernel headers and sources for building any Linux kernel module. The package(s) are typically named kernel-devel or kernel-sources / kernel-headers. If a kernel was custom built from http://kernel.org/ sources, the kernel build tree must be available for the Violin Linux driver to reference.

GCC Compiler The GCC Compiler compiles the Linux kernel and modules. The GCC Compiler version should match the version that was used to build the running kernel.

Libaio Development Package The Libaio Development Package is an asynchronous I/O development package for applications that use libaio.so. The perf_test benchmark utility (included among the Violin utilities) uses the asynchronous I/O features to provide high performance asynchronous I/O benchmarking. See Performance Tuning and Testing on page 423 for more information on the perf_test utility and asynchronous I/O testing. Install both the libaio and libaio-devel packages before installing the Violin Utilities. The following example installs the required packages on a CentOS 5.5 x86_64 system. At the root prompt, type: # yum install gcc kernel-devel-`uname -r`.x86_64 libaio-devel

Installing Violin Linux Drivers The second step towards configuring a direct-attached Violin Memory Array system with a Linux host is to install the Violin Linux driver. The Violin Linux driver enables a direct-attached PCIe connection between the Violin Memory Array and the Linux host machine. In the course of installing the Violin Linux driver, you must decide whether the Violin Linux driver is to be loaded manually or loaded automatically as a module when the host boots up. This section describes the installation and removal of the Violin Linux drivers:

• •

Installing Violin Linux Drivers on page 390 Uninstalling Violin Linux Drivers on page 393

535-0008-00 Rev 10


389


Installing Violin Linux Drivers Before installing the Violin Linux driver, ensure that the host computer has the kernel-devel RPM installed and GCC installed for the running kernel, which will allow the driver source to compile. These packages may not be installed by default on the Linux host. The information shown in the following examples may vary from your installation depending on the host computer kernel and distribution. The Violin Linux driver may be installed on either a Linux kernel that is running or not running. Both procedures are documented in this section.

•

Installing Violin Linux Drivers on Running Linux Kernels on page 390

Installing Violin Linux Drivers on Running Linux Kernels To load the Violin Linux driver onto the host computer: 1. Power up the Memory Array and verify the PCIe connectivity between the Memory Array and

the host computer before you install and load the Violin Linux driver onto the host. Because the PCIe Hot Plug is not supported by most current BIOS versions, you must power up the Memory Array before powering up the host computer. 2. Log into the host computer as root. 3. Download the Violin Linux install file (vtms-linux-driver.run) from the Violin

Memory Customer Support ftp site. Where indicates the release number; for example, D6.0.0.3 4. Save the Violin Linux install file to a suitable location on the host. 5. Open a terminal shell window and navigate to the location of the Violin Linux install script. 6. To run the Violin Linux install script, type:

./vtms-linux-driver-.run

390


535-0008-00 Rev 10


The script begins installing the driver, as shown in the following example: # ./vtms-linux-driver-D6.0.0.3.run Copying to a temporary location... Creating directory vtms-linux-driver-D6.0.0.3 Verifying archive integrity... All good. Uncompressing Violin Memory, Inc. Linux Driver source.............................. Current settings detected: ------------------------------------------------Kernel = 2.6.32-100.28.5.el6.x86_64 Kernel Dir = /lib/modules/2.6.32-100.28.5.el6.x86_64/build Kernel GCC = 4.4.4 GCC ver = 4.4.4 Arch = x86_64 View the driver README.txt? [n]: Skipping /tmp/vtms-linux-driver-D6.0.0.3/README.txt 7. Type y if you want to view the README.txt.

535-0008-00 Rev 10


391


8. When prompted to build the Violin Linux driver type y: Build the driver now? [y]: Detected kernel build dirs: ------------------------------------------------/lib/modules/2.6.32-100.28.5.el6.x86_64/build /lib/modules/2.6.32-71.el6.x86_64/build /lib/modules/2.6.35.14/build /lib/modules/2.6.35.7/build /lib/modules/3.0.0/build Build dir [/lib/modules/2.6.32-100.28.5.el6.x86_64/build]: Building driver for 2.6.32-100.28.5.el6.x86_64 make[1]: Entering directory `/tmp/vtms-linux-driver-D6.0.0.3' make[1]: Leaving directory `/tmp/vtms-linux-driver-D6.0.0.3' make[1]: Entering directory `/usr/src/kernels/2.6.32-100.28.5.el6.x86_64' CC [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms_strad.o CC [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms_hvm.o CC [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms_strad_cmn.o CC [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms_smart.o CC [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms_strad_bdev.o CC [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms_strad_cdev.o CC [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms_strad_scsi.o LD [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms.o Building modules, stage 2. MODPOST 1 modules CC /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms.mod.o LD [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms.ko.unsigned NO SIGN [M] /tmp/vtms-linux-driver-D6.0.0.3/2.6.32-100.28.5.el6.x86_64/ x86_64/vtms.ko make[1]: Leaving directory `/usr/src/kernels/2.6.32-100.28.5.el6.x86_64' Build complete. 9. At the prompt, you are asked if you want to install the Violin Linux driver in the /lib/modules

directory.

•

If you select yes, the Violin Linux driver module is added to the kernel modules.

•

If you select no, the Violin Linux driver is not installed to the /lib/modules directory.

A message appears stating that the Violin Linux driver installation is complete. 10.Once the installation is complete you may load the Violin Linux driver for use.

392


535-0008-00 Rev 10


•

If you installed the Violin Linux driver in the /lib/modules directory, see Loading Violin Linux Drivers as Modules on page 395 for detailed instructions.

•

If you did not install the Violin Linux driver in the /lib/modules directory, see Loading Violin Linux Drivers Manually on page 394.

Preventing Violin Linux Drivers from Loading on Boot Up To prevent the kernel from loading the Violin Linux driver at bootup, add the Violin Linux driver to the /etc/modprobe.d/blacklist file. Type modprobe vtms once the system is booted to load the driver. Ensuring that the Violin Linux Driver Loads on Boot Up Depending on your Linux distribution, you may need to modify the /etc/modprobe.conf file to include an alias for the Memory Array if you want the Violin Linux driver to automatically load on bootup. To add an alias for the Violin Linux driver, edit the file such as the following: alias scsi_hostadapterN vtms Where N specifies the next available integer or blank if it is the first alias entry for a SCSI host adapter.

Uninstalling Violin Linux Drivers 1. Log into the Linux host as root. 2. Open a terminal window and at the system prompt, type:

cd /tmp/vtms-linux-driver- make uninstall The Violin Linux driver is removed from the /lib/modules directory, but can still be loaded and unloaded using the steps described in Loading and Unloading Linux Drivers on page 394. Load the Violin Linux driver again if you want to use the Memory Array. See Installing Violin Linux Drivers on page 390 for details on installing the Violin Linux driver.

535-0008-00 Rev 10


393


Loading and Unloading Linux Drivers The third step towards configuring a direct-attached Violin Memory Array system with a Linux host is to load the Violin Linux driver. Violin Linux drivers may be configured to be loaded automatically as modules by the Linux host or to require manually loading. This section describes the loading and unloading of Violin Linux drivers in three topics:

• • •

Loading Violin Linux Drivers Manually on page 394 Loading Violin Linux Drivers as Modules on page 395 Unloading the Violin Linux Driver on page 395

Loading Violin Linux Drivers Manually If you did not install the driver in the /lib/modules directory, you must reload the driver manually after each system reboot. To load the Violin Linux driver manually: 1. Power up the Memory Array and verify the PCIe connectivity between the Memory Array and

the host computer before installing and loading the Violin Linux driver onto the host computer. 2. Log into the host computer as root. 3. At the system prompt, navigate to the following directory:

cd /tmp/vtms-linux-driver-// Where:

Indicates the release number, such as D6.0.0.3

Indicates the kernel version, such as 2.6.9-42.ELsmp.

Indicates the architecture type of the host computer, such as x86_64.

4. Load the Violin Linux driver; at the system prompt, type:

./load_strad.sh 5. At any point, if you decide to install the Violin Linux driver in /lib/modules at the system

prompt, type: cd /tmp/vtms-linux-driver- make install depmod

394


535-0008-00 Rev 10

Loading and Unloading Linux Drivers

Where: Indicates the release number, such as 4.5.4.1.

Loading Violin Linux Drivers as Modules To load the Violin Linux drivers as modules: 1. Power up the Memory Array and verify the PCIe connectivity between the Memory Array and

the host computer before installing and loading the Violin Linux driver onto the host computer. 2. Log into the Linux host as root. 3. Open a terminal window. 4.

(Optional) To update the dependency mappings of newly installed modules, at the system prompt, type: depmod You may need to run depmod for modprobe to detect the new module.

5. To load the Violin Linux driver, at the system prompt, type:

modprobe vtms [] Where indicates device type of the loaded driver. Valid values are: none

Load as a block device

[use_cdev=1]

Load as a character device

[use_scsi=1]

Load as a SCSI device

After loading the Violin Linux driver, create partitions and file systems on the Memory Array or directly access the character device.

Unloading the Violin Linux Driver 1. Log into the host computer as root. 2. Open a terminal window. 3. To unload the Violin Linux driver, at the system prompt, type:

rmmod vtms

535-0008-00 Rev 10


395


Installing Violin Utilities The fourth step towards configuring a direct-attached Violin Memory Array system with a Linux host is to install the Violin utilities. Violin utilities are tools designed to monitor the performance of a direct-attached Memory Array system. Using Violin utilities, you may retrieve detailed information about the Memory Array and data transfer counts that enable you to configure the system for optimal performance. Use the Violin utility commands in the /usr/local/bin directory to display information on the Memory Array.

Utility

Description

perf_test

Violin multi-threaded AIO-capable disk benchmark utility. See Performance Testing Using Perf_Test on page 428 for detailed instructions on using this utility.

vcounts

Displays Memory Array I/O counters.

veeprom

Displays information, such as the main board serial number and the management MAC address.

vinfo

Displays driver version and registry tunable parameters.

vpartial

Displays Memory Array partial Flash page counters.

vstat

Displays Memory Array status information.

vzero

Resets Memory Array I/O counters to zero.

Table C.1 Memory Array Commands for Viewing Violin Linux Driver Information

For the complete syntax and usage details of each command, see Violin Utilities on page 403. On each command, specify the optional device index ( [ ] ) to list the information for a particular Memory Array. If you do not specify an individual device, the command lists the information for all the Memory Arrays found.

Installing Violin Utilities on Linux Hosts To install the utilities onto the host computer: 1. Log into the host computer as root. 2. Download the Violin Utilities Package tgz file (vtms-linux-utils-.tgz) from

the Violin Memory Customer Support ftp site. Where indicates the release number, such as D6.0.0.3 3. Extract the contents of the compressed file to a suitable location on the host. 4. Open a terminal shell window and navigate to the location of the Violin Utilities install script

(vtms-linux-.tar).

396


535-0008-00 Rev 10

Configuring Partitions and Storage Devices

5. To run the Violin utilities install script, type:

./vtms-linux-utils-.run Where indicates the release number, such as D6.0.0.3

Configuring Partitions and Storage Devices In a direct-attached Violin Memory Array system with a Linux host, block device partitions must be configured. Create block device partitions on the Memory Array for creating multiple file system partitions. Use the /dev/vtmsa device for creating file systems only if a single large partition is required. When creating partitions for 4kB partition alignment, the fdisk utility has a -S option that lets you override the default geometry of 63 sectors per track. Use fdisk with the -S56 option for 4kB alignment on created partitions. For more information, see Memory Array Flash 4kB Alignment on page 425.

Creating Block Device Partitions To create a block device partition on the Memory Array: 1. Log into the host computer as root.

The following command is only needed when splitting the block device into multiple partitions. 2. First, open the device for adding partitions by typing the following at the command prompt:

fdisk –S56 /dev/vtmsa Where /dev/vtmsa specifies the device to open. The information shown below is returned and you will be in the block device setup menu. # fdisk –S56 /dev/vtmsa The number of cylinders for this disk is set to 16709. There is nothing wrong with that, but this is larger than 1024, and could in certain setups cause problems with: 1) software that runs at boot time (e.g., old versions of LILO) 2) booting and partitioning software from other OSs (e.g., DOS FDISK, OS/2 FDISK) Command (m for help): 3. At the Command prompt, type n to create a new partition:

n

535-0008-00 Rev 10


397


Where n specifies to create a new partition. The command displays information similar to the following: Command (m for help): n Command action e extended p primary partition (1-4)

4. Next, at the Command prompt, create the new partition by typing:

p Where p creates a primary partition. The command displays information similar to the following: Command action e extended p primary partition (1-4) p Partition number (1-4): 1 First cylinder (1-16709, default 1): Using default value 1 Last cylinder or +size or +sizeM or +sizeK (1-16709, default 16709): 8000

•

Specify the partition number if you will be creating more than one partition on the device, or accept the default value of 1 by pressing the Enter key. The default value to set the beginning cylinder to is 1, which can be changed if needed.

•

Specify the size of the block device or accept the default value of the available memory by pressing the Enter key.

5. Next, at the Command prompt, write the new partition by typing:

w Where w writes the partition table to disk and exit. The command displays information similar to the following: Command (m for help): w The partition table has been altered! Calling ioctl() to re-read partition table. Syncing disks. #

The system command prompt is returned. Finish by creating the file system that uses the new block device partition, creating the new mount path, and mounting the new block device file system. 398


535-0008-00 Rev 10


6. Create a file system on the block device by typing:

mkfs -t ext3 /dev/vtmsa Where: -t ext3

Specifies the type of file system to be built.

/dev/vtmsa

Specifies the name of the new file system, where is the assigned device number.

The command displays information similar to the following: # mkfs -t ext3 /dev/vtmsa1 mke2fs 1.39 (date) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) 176160768 inodes, 352321536 blocks 17616076 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=4294967296 10752 block groups 32768 blocks per group, 32768 fragments per group 16384 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968, 102400000, 214990848 Writing inode tables: done Creating journal (32768 blocks): done Writing superblocks and filesystem accounting information: done This filesystem will be automatically checked every 37 mounts or 180 days, whichever comes first. Use tune2fs -c or -i to override. 7. Create the new mount path for mounting the block device by typing:

mkdir /mnt/strad Where /mnt/strad specifies the name of the new mount path. 8. At the command prompt, mount the new file system by typing:

mount –t ext3 /dev/vtmsa /mnt/strad

535-0008-00 Rev 10


399


Where: -t ext3

Specifies the type of file system to mount.

/dev/vtmsa

Specifies the name of the new file system, where is the assigned device number.

/mnt/strad

Specifies the name of the new mount path.

9. To unmount the file system, at the command prompt, type:

umount /mnt/strad

Enabling Host to Automatically Mount Partitions By editing the /etc/fstab file, you can enable the host computer to automatically mount Memory Array partitions on bootup. 1. Log into the host computer as root and change directories into /etc. 2. Edit the /etc/fstab file with the desired text editing tool. 3. Add the following row to the file, changing the values to appropriate ones for your system: /dev/ /mnt/

Where: /dev/ /mnt/

400

Specifies the partition name. Specifies the default mount path. Specifies the file system type of the device. Specifies the mount options, which are separated by commas, for the file system. The available options are: auto | noauto—Specifies whether to automatically mount the device or not. With noauto, the device can only be mounted explicitly. user | nouser —Specifies whether a typical user versus the “root” user can mount the device. exec | noexec —Specifies that the binaries contained on the partition can or cannot be executed. Not recommended for a root partition. ro—Mounts the device read-only. rw—Mounts the device read-write. sync | async —Specifies how input and output is performed by the device. For example, a command is executed at the same time it has been issued. defaults —Specifies to use the default options, which are: rw, suid, dev, exec, auto, nouser, async.


535-0008-00 Rev 10


Specifies whether the device is backed up by the dump utility or not. If 0 is specified, the device will not be backed up. 1 specifies that the device will be backed up. Specifies what order the devices are checked. If 0 is specified, the device won’t be checked by fsck.

Enabling Logical Volume Manager (LVM) to Recognize Direct-Attached Storage By default, Logical Volume Manager (LVM) does not correctly recognize the Violin 6000 Series Memory Array in a direct-attached configuration. Block devices named /dev/vtms(a-z) are created when installing the Violin Linux driver. LVM only scans for certain device names when looking for devices to manage, and by default, vtms(a-z) is not scanned. This issue can be resolved by installing a patch included with the Linux driver installation. The patch modifies the /etc/lvm/lvm.conf file so that it recognizes the Violin 6000 Series Memory Array block device files. 1. Log into the host computer as root. 2. Within the Linux driver tar (.tgz) file, expand the file vtms-linux-app-patch-

.tgz, where indicates the release number; for example, D6.0.0.3. The expanded file contains the following two directories:

• •

lvm openfiler

3. Open the lvm directory. 4. Run the installation script titled vtms-lvm-patch.sh on the Linux system by typing the

following: sudo ./vtms-lvm.patch.sh

535-0008-00 Rev 10


401


402


535-0008-00 Rev 10

APPENDIX D

Violin Utilities

This chapter describes Violin utilities for configuring and performance testing the Violin 6000 Series Memory Array system. These commands are common to the Violin Linux driver and to the Violin Memory Gateway.

• •

Understanding Violin Utilities on page 403 Violin Utilities Reference on page 405

Understanding Violin Utilities Violin utilities are tools designed to monitor the performance of Memory Array systems in a direct-attached configuration. Using Violin utilities, you may retrieve detailed information about the Memory Array and data transfer counts that enable you to configure the system for optimal performance.

Installing Violin Utilities The Violin utilities may be installed on the Linux host when the Violin drivers are installed for a direct-attached Memory Array deployment. See Violin Drivers on page 387 for more information.

Running Violin Utilities As with any command that modifies the configuration of the Memory Array system, Violin utility commands must be issued as the root user on a Linux computer or as an Administrator on a Windows computer. On each Violin utility command, specify the optional device index ([]) to list the information for a particular Memory Array. If you

535-0008-00 Rev 10


403

Appendix D - Violin Utilities

do not specify an individual device, the command lists the information for all Memory Arrays found. Running Violin Utilities on a Linux Host Violin utilities are run on the Linux host from the command line. Running Violin Utilities on a Windows Host A shortcut icon is created on the Windows desktop to the Violin Utilities command shell when you install the Violin Utilities on the Windows host. If you are not logged on as Administrator, right-click the icon and then select Run As Administrator to open the shell with the proper permissions. Running Violin Utilities from the Memory Array Violin utilities must be run from the Master Memory Gateway. The following is an example of the steps required to log in to the internal Master Memory Gateway on a Violin 6000 Series Memory Array. 1. Start a terminal application and log in to the Master ACM as “admin”. The Master ACM is set

up during initial configuration of the Violin 6000 Series Memory Array. Enter the system host name or the ACM management IP address to ensure a direct connection. 2. Once logged in to the Master ACM, type enable at the command prompt:

violin-acma [violin: master] > enable█ 3. Log in to the Master Memory Gateway by typing the following command:

violin-acma [violin: master] # slogin mg-master█ 4. At the command prompt, type enable:

violin-mga [violin: master] > enable█ You can now run the Violin utilities.

404


535-0008-00 Rev 10

Violin Utilities Reference

Violin Utilities Reference This section documents the syntax and usage of the following Violin utilities.

• • • • • • • • • • • • •

varray on page 405 vcounts on page 406 veeprom on page 409 vincident -a on page 410 vinfo on page 410 vinventory on page 412 vmesg on page 413 vpartial on page 413 vring on page 415 vspeedtest on page 415 vstat on page 416 vvimms on page 419 vzero on page 419

varray The varray command displays array information, including number of VIMMs, temperature, fan speed and LED states. Syntax varray

535-0008-00 Rev 10


405

Appendix D - Violin Utilities

Example The varray utility returns information similar to the following when run on a Linux host: violin-mga [violin: master] (config) # varray Violin Memory, Inc. Version: vtms-linux-utils-V6.0.0 03/13/2013

-- Memory Array Virtual Device-Device: /dev/vtmsa Index: 0 -- Memory Array -Chassis Type: Number of VIMMs: Ambient Temperature: Controller Temperature: Power A: Power B: Uptime: Lid Ajar Time: Fan 0: Fan 1: Fan 2: Fan 3: Fan 4: Fan 5: Alarm LED: Status LED: Power A LED: Power B LED:

V6000 64 20C 42C ON ON 242745 secs 0 secs Slow Slow Slow Slow Slow Slow OFF ON ON ON

vcounts The vcounts utility displays data transfer counters for the Memory Array. Syntax vcounts [ ]

406


535-0008-00 Rev 10

Violin Utilities Reference

Example The vcounts utility returns information similar to the following when run on a Linux host: # vcounts Violin Memory, Inc. Version: vtms-linux-utils-V6.0.0 03/13/2013 Device: Index:

/dev/vtmsa 0

-- Target Counts IO > 128K 64K < IO Information page. Each file is described in the table below.

This file...

Provides this information...

VIOLIN-MEMORY-ROOT-MIB.txt

The root Violin Memory Array and Gateway Object Identifiers (OIDs).

VIOLIN-MEMORY-ARRAY-MIB.txt

Individual Memory Array module information, including, model and serial numbers, software version, temperature, VIMM health, format, read and write, link speed, power and other states.

VIOLIN-MEMORY-MEDIA-MIB.txt

Basic Memory Array information, including chassis information, module states, alarms, VIMM life, format percentage and state, IP addresses, LED information, I/O information, read and write operations, and more.

VIOLIN-MEMORY-TRAP-MIB.txt

OIDs for SNMP traps sent from the Memory Array to configured destination points to alert when events occur, such as the lid is ajar, state changes to license keys, fans, and more.

TallMaple-MIB.txt

Basic system information about CPU load, process failures, mounted file systems, and more.

SNMP MIB Files

Below are example URLs for MIBs on a Violin Memory Array named HOSTNAME: http://HOSTNAME/doc/VIOLIN-MEMORY-MIB.txt http://HOSTNAME/doc/TallMaple-MIB.txt Traps: http://HOSTNAME/doc/VIOLIN-MEMORY-TRAP-MIB.txt

Benefits of Using a MIB Browser You can obtain a free MIB Browser from iReasoning (Linux, Mac, Windows) that allows a load of up to 10 MIBs. An MIB Browser is a tool for managing SNMP enabled network devices and applications. It allows you to load MIBs, issue SNMP requests to retrieve agent's data, or make changes to the agent. A built-in trap receiver can receive SNMP traps and handle trap storm. To download a free iReasoning MIB Browser, go to the company Web site at http://ireasoning.com

434


535-0008-00 Rev 10

SNMP Configuration on the Violin 6000 Series Memory Array

SNMP Configuration on the Violin 6000 Series Memory Array Use the snmp and snmp-server commands to configure SNMP. It defaults to on. violin-acma [violin: SNMP enabled: SNMP port: System location: System contact: Read-only community: Traps enabled: Trap community:

master] (config) # show snmp yes 161

public yes public

Interface listen enabled: yes No Listen Interfaces. No trap sinks configured. violin-acma [violin: master] (config) #█

535-0008-00 Rev 10


435

Appendix F - SNMP Traps

Trap/Notify Event E-mail and Logged Events This section lists the trap notify settings, along with their respective event e-mail subject lines, classifications, and logged events.

Trap/Notify Setting user-login user-logout process-exit liveness-failure cpu-util-high cpu-util-ok

Event Email Subject

User %s logged in%s User %s logged out%s Process exit: %s Process hung: %s CPU utilization too high CPU utilization alarm clearing paging-high Paging activity too high paging-ok Paging activity alarm clearing disk-space-low Disk space low disk-space-ok Disk space OK free-space-low Low free space memusage-high Memory utilization too high memusage-ok Memory utilization OK netusage-high Network utilization too high netusage-ok Network utilization OK disk-io-high Disk I/O too high disk-io-ok Disk I/O OK unexpected-cluster-join A node joined the cluster unexpectedly unexpected-cluster-leave A node left the cluster unexpectedly unexpected-cluster-size Unexpected cluster size event on %s — Disk failure warning: device %s cache-faults A cache fault was detected hwmon-file-sys-error File system malfunction unexpected-shutdown Unexpected shutdown interface-down %s link down interface-up %s link up hwmon-mce-error Machine check exception media-device-healthMedia device health warn check warning media-device-stateUnexpected device change removal media-device-stateUnexpected device change state change

436

Class.

Logged Event

Event Event Event Event Event Event

/mgmtd/session/events/login /mgmtd/session/events/logout /pm/events/failure/unexpected_exit /pm/events/failure/liveness /stats/events/cpu_util_indiv/rising/error /stats/events/cpu_util_indiv/rising/clear

Event Event

/stats/events/paging/rising/error /stats/events/paging/rising/clear

Event Event Event Event

/stats/events/fs_mnt_bytes/falling/error /stats/events/fs_mnt_bytes/falling/clear /vshare/events/array/free_space_trigger /stats/events/memory_pct_used/rising/error

Event Event

/stats/events/memory_pct_used/rising/clear /stats/events/intf_util/rising/error

Event Event Event Event

/stats/events/intf_util/rising/clear /stats/events/disk_io/rising/error /stats/events/disk_io/rising/clear /cluster/events/unexpected_join

Event

/cluster/events/unexpected_leave

Event

/cluster/events/unexpected_size

Failure /smart/events/warning Event

/crd/events/internal_fault

Failure Failure Event Event Event

/hwmon/events/failure/filesystem /system/events/unexpected_shutdown /net/interface/events/link_down /net/interface/events/link_up /hwmon/events/failure/mce

Event

/media/events/device/health_warning

Event

/media/events/device/state_change

Event

/media/events/device/state_change


535-0008-00 Rev 10

Trap/Notify Event E-mail and Logged Events

media-device-unknowntype media-device-lifetimewarn media-system-swapstate media-system-swapstate excessive-shutdowns

Unknown media device detected Cache device lifetime warning System swap device activated System swap disabled

Excessive unexpected shutdowns license-state-change A license for %s %s conntrack-entries-high Too many connection tracking entries conntrack-entries-ok Connection tracking entries alarm clearing vimm-temperature-high VIMM temperature alarm detected vimm-temperature-ok VIMM temperature alarm clearing chassis-temperature-high Chassis temperature alarm detected chassis-temperature-ok Chassis temperature alarm clearing lid-ajar-time-high Excessive lid ajar time detected lid-ajar-time-ok Lid ajar alarm clearing lid-ajar Array lid ajar state change detected array-pcie-link-up Array PCIE link up detected array-pcie-link-down Array PCIE link down detected array-data-plane-ready Array data plane ready state change detected array-raid-rebuild Array RAID rebuild state change detected vimm-state-change Array VIMM state change detected array-psu-state Array PSU state change detected array-led-change Array LED state change detected array-fan-change Array FAN state change detected fc-port-state-change vSHARE FC port state change detected ib-port-state-change vSHARE IB port state change detected vmg-fan-change Gateway FAN state change detected

535-0008-00 Rev 10

Event

/media/events/device/unknown_device_type

Event

/media/events/device/life_remain_warning

Event

/media/events/system/swap_state_change

Event

/media/events/system/swap_state_change

Failure /shutdown_check/events/ excessive_shutdowns Event /license/events/key_state_change Event /stats/events/conntrack_entries/rising/error Event

/stats/events/conntrack_entries/rising/clear

Event

/media/events/vimm_temp/rising/error

Event

/media/events/vimm_temp/rising/clear

Event

/media/events/chassis_temp/rising/error

Event

/media/events/chassis_temp/rising/clear

Event

/media/events/lid_ajar_time/rising/error

Event Event

/media/events/lid_ajar_time/rising/clear /media/events/array/lid_ajar

Event

/media/events/array/pcie_link_up

Event

/media/events/array/pcie_link_down

Event

/media/events/array/data_plane_ready

Event

/media/events/array/raid_rebuild

Event

/media/events/array/vimm_state_change

Event

/media/events/array/psu_state

Event

/media/events/array/led_change

Event

/media/events/array/fan_change

Event

/vshare/events/fc/port_state_change

Event

/vshare/events/ib/port_state_change

Event

/vshare/events/vmg/fan_change


437


vmg-psu-change vmg-boot-raid-change

Gateway PSU state change detected Gateway Boot Disk RAID state change detected

Event

/vshare/events/vmg/psu_change

Event

/vshare/events/vmg/boot_raid_change

Examples Example Failure Email Subject System failure on : Unexpected shutdown Example Event Email Subject System event on : Unexpected cluster size event on Example Event Log May 25 01:44:17 njvshare mgmtd[4148]: [mgmtd.INFO]: EVENT: /vshare/events/fc/port_state_change

Available Traps The following section covers the available traps. The first table lists the traps that are enabled by default. The second table lists the traps that are disabled by default, and that you can enable as needed.

438


535-0008-00 Rev 10

Available Traps

Default Enabled Traps New Traps in Violin Memory Gateway Release G5.2.0

Traps

Description

ib-port-state-change

InfiniBand port state change detected

lid-ajar

Lid ajar alarm

vmg-boot-raid-change

VMG-350 and HP DL380 only. Boot disk status changed.

vmg-fan-change

VMG-3xx and HP DL380 only. Fan state changed.

vmg-psu-change

VMG-3xx and HP DL380 only. Power supply status changed.

New Traps in Violin Memory Gateway Release G5.1.x

Traps

Description

array-data-plane-ready:

Array data plane state changed.

array-led-change:

Array LED state changed.

array-pcie-link-down:

Array PCIE link down.

array-pcie-link-up:

Array PCIE link up.

array-psu-state:

Array PSU state changed.

array-raid-rebuild:

Array RAID rebuild state changed.

chassis-temperature-high: High chassis temperature detected. chassis-temperature-ok:

Chassis temperature returned to normal levels.

lid-ajar-time-high:

Excessive lid ajar time detected.

lid-ajar-time-ok:

Lid ajar alarm cleared.

vimm-state-change:

Array VIMM state changed.

vimm-temperature-high:

High VIMM temperature detected.

vimm-temperature-ok:

VIMM temperature returned to normal levels.

Traps Prior to Violin Memory Gateway Release G5.1.x

Traps

Description

cpu-util-high:

CPU utilization has risen too high.

disk-space-low:

File system free space has fallen too low.

interface-down:

An interface's link state has changed to down.

interface-up:

An interface's link state has changed to up.

liveness-failure:

A process in the system was detected as hung.

paging-high:

Paging activity has risen too high.

process-crash:

A process in the system has crashed.

process-exit:

A process in the system unexpectedly exited.

535-0008-00 Rev 10


439


snmp-authtrap:

An SNMP v3 request has failed authentication

unexpected-shutdown:

Unexpected system shutdown.

Default Disabled Traps An additional 16 SNMP traps are available that are disabled by default.

Trap

Description

array-fan-change:

Array FAN state changed.

disk-io-high:

Disk I/O per second has risen too high.

fc-port-state-change:

vSHARE FC port state changed.

license-state-change:

A license has changed state.

media-device-healthwarn:

A media device has a health status warning.

media-device-lifetimewarn:

A media device has a low est. life remaining.

media-device-statechange:

Unexpected change in media device state.

media-device-unknowntype:

Detected a media device of unknown type.

media-system-swapstate:

Detected a change in system swap state.

memusage-high:

Memory usage has risen too high.

netusage-high:

Network utilization has risen too high.

unexpected-cluster-join:

A node has unexpectedly joined the cluster.

unexpected-clusterleave:

A node has unexpectedly left the cluster.

unexpected-cluster-size:

The number of nodes in the cluster is unexpected.

user-login:

A user has logged into the system.

user-logout:

A user has logged out of the system.

Trap Information Table

Trap Name

array-data-planeready

440

Threshold

Value Unit

State Change 0 or 1

Description

A Violin Array data plane available state change has occurred. The old value and new values are supplied. A value of 0 = not ready, 1 = ready.


Solution

If the new value is 0, then check the Violin Array alarms for more information.

Sev.

High / Low

535-0008-00 Rev 10

Available Traps

Trap Name

Threshold

Value Unit

Description

Solution

Sev.

array-fan-change

State Change State

One or more fans have changed state on a Violin Array. The old and new values are provided using the following states: OFF, Absent, Low, Medium, or High.

Check Violin Array for alarms as well as airflow and operating environment.

Medium

array-led-change

State Change State

One or more of the LEDs on a Violin Array has changed state. The following LEDs are tracked for state change: Alarm, PowerA, PowerB, and Status. The LED values are: OFF, SLOW_BLINK, FAST_BLINK, or ON.

If the new value is OFF for either Power LED, then check power cables. If Status LED is anything but ON, or Alarm LED is anything but OFF, then check the Violin Array for alarms.

High / Low

array-pcie-link-down

Event

Triggered

The Violin Array specified in the trap has detected a PCIe link loss.

Check the PCIe cable between the Violin Array and the Violin Memory Gateway.

High

array-pcie-link-up

Event

Triggered

The Violin Array specified in the trap has detected that the PCie link has gone active.

Nothing.

Low

array-psu-state

State Change State

A Violin Array has changed state regarding one or both of the Power Supply Units. A true or false value for psuA and psuB specifies whether the old / new values are different. The PSU state can be: Absent, ON, or OFF.

Check the power source to the Violin Array power supplies.

High / Low

535-0008-00 Rev 10


441


Trap Name

Threshold

Value Unit

Description

Solution

Sev.

array-raid-rebuild

State Change 0 or 1

A Violin Array has changed state regarding VIMM RAID group rebuild. A new value of 1 specifies that a RAID rebuild is in progress and that performance will be affected. A new value of 0 specifies that the RAID rebuild has completed.

If the new value is 1, check the Violin Array alarms for a failed VIMM and contact Violin support for a possible replacement.

Medium

chassis-temperaturehigh

75

Celsius

Temperature inside a Violin Array chassis has exceeded normal operating range

Check airflow and operating environment.

High

chassis-temperatureok

70

Celsius

Temperature inside a Violin Array chassis has dropped into normal operating range

Nothing.

Low

cpu-util-high

98

Percent

A Violin Memory Gateway has detected that a CPU has exceeded utilization above the threshold level

Please contact Violin customer support if the system is not under full data traffic load.

Low

disk-io-high

5120

KBytes / sec

The internal disk on a Violin Memory Gateway has crossed a threshold for performing too many I/ O’s per second.

If the system is not under data traffic load, please contact Violin customer support.

Low

disk-space-low

0

Percent Free

A Violin Memory Gateway disk space has crossed the threshold of percent of bytes free.

Please contact Violin customer support.

Medium

442


535-0008-00 Rev 10

Available Traps

Trap Name

Threshold

Value Unit

Description

Solution

Sev.

fc-port-state-change

State Change State

One or more Fibre Channel ports have changed state on a Violin Memory Gateway. The old and new values for state are provided from one of: Unknown, Failover Failed, Failover, Not Supported, Online, Lost, Dead, Unconfigured. The FC port speed is included in the data but not used to generate the state change. The following speed values are used: 1 Gbit, 2 Gbit, 4 Gbit, and 8 Gbit.

Verify that OM-3 rate Fibres are used in your config for 8 Gb. Check that the Fibres and SFP ports were properly cleaned and that the cables and SFP’s are fully inserted. Verify that the Fibre does not exceed bend radius specifications.

Medium

free-space-low

User configurable

Available free space on the array fell below the threshold.

Adjust the free space threshold to activate space reclamation, and/or delete unused snapshots.

Medium

ib-port-state-change

State Change Active, Down, Initialize, Armed

vSHARE InfiniBand (IB) port state changed.

One or more of the Medium to High IB ports have changed state on a Violin Memory Gateway. The old and new state values are provided, as well as the IB port speed. If a port is down, contact Violin Memory Customer Support.

interface-down

Event

Triggered

A network interface on a Violin Memory Gateway has lost link.

Check the network cables to the Violin Memory Gateway.

High

interface-up

Event

Triggered

A network interface on a Violin Memory Gateway has detected link up.

Nothing.

Low

535-0008-00 Rev 10

Percent Free


443


Trap Name

Threshold

Value Unit

Description

Solution

Sev.

license-state-change

State Change State

A Violin Memory Gateway feature license state has changed. The old and new active state is passed as well as the specific feature license. The active state is a “true” or “false” value. When true, the feature license is active.

If a license has changed to active = false, please contact Violin customer support to see if your license has expired.

Medium

lid-ajar

State Change Triggered

Array lid is open.

Close and secure the lid.

High

lid-ajar-time-high

60

Seconds

A Violin Array chassis lid is open

Verify Violin Array High chassis lid is closed

lid-ajar-time-ok

1

Seconds

A Violin Array chassis lid has been closed for at least this period

Nothing.

Low

liveness-failure

Event

Triggered

An internal process on a Violin Memory Gateway has been detected as hung.

Please contact Violin customer support

High

media-device-healthwarn

10

Percent

Check the Violin A Violin Memory Gateway media device Array for any (Violin Array) has one alarms. or more health attributes in a warning or critical state. Normalized S.M.A.R.T attributes are used. The key attributes for Violin Arrays are Avail Reserved Space and Media Wearout Indicator. S.M.A.R.T attr 232: Avail Reserved Space is the percentage of reserved blocks available (unused). This threshold is 10%. S.M.A.R.T attr 233: Media Wearout Indicator tracks the number of erase cycles for flash as a percentage of life remaining.

444


High

535-0008-00 Rev 10

Available Traps

Trap Name

Threshold

Value Unit

Solution

Sev.

A Violin Memory Gateway media device (Violin Array) has crossed the threshold for S.M.A.R.T attr 233: Media Wearout Indicator. This tracks the number of erase cycles for flash as a percentage of life remaining.

Contact Violin customer support to determine if any VIMMs should be replaced.

High

media-devicelifetime-warn

5

media-device-statechange

State Change State

A Violin Memory Gateway media device (Violin Array) has changed state. The possible states are: unknown, online, offline, error, removed, disabled, or onlining.

Verify that the connected Violin Array is in the proper online state.

High / Low

media-deviceunknown-type

Event

A Violin Memory Gateway has detected a media device connected of an unknown type. The Model, Serial #, and size are provided.

The most common occurrence is a freshly formatted Violin Array has been connected to a Violin Memory Gateway. The CLI “media init ...” command is used to initialize the device.

Low

media-system-swapstate

State Change State

An internal Memory Gateway disk has changed state for having an active swap partition. The active state is true when a swap partition has been activated or false for deactivation.


Low

memusage-high

90

A Violin Memory Gateway has detected that system RAM has exceeded a percentage in use threshold.


Medium

535-0008-00 Rev 10

Percent

Description

Triggered

Percent


445


Trap Name

Threshold

Value Unit

Description

Solution

Sev.

netusage-high

10485760

Bytes / sec A Violin Memory Gateway has detected that a network interface has exceeded a bytes per second threshold.

Please verify your network configuration. For 10 GbE iSCSI, the threshold should be adjusted much higher.

Low

paging-high

2000

Duration / sec

Memory paging on a Violin Memory Gateway has exceeded a threshold.

Please contact Violin customer support if the system is not under full data traffic load.

Medium

process-crash

Event

Triggered

An internal process on a Violin Memory Gateway has crashed. A callhome event is generated with the details of the event.


High

process-exit

Event

Triggered

An internal process on a Violin Memory Gateway has unexpectedly exited. A callhome event is generated with the details of the event.


High

unexpected-clusterjoin

Event

Triggered

A Violin Memory Gateway has unexpectedly joined a vCLUSTER after the 180 seconds cluster startup time has elapsed.

Check the Violin Memory Gateway log for any errors. Please contact Violin customer support if the join event is unexplained.

Low

unexpected-clusterleave

Event

Triggered

A Violin Memory Gateway has unexpectedly left a vCLUSTER after already being a member.

Check power and network connectivity to the missing VMG. Please contact Violin customer support if the leave event is unexplained.

High

446


535-0008-00 Rev 10

Available Traps

Trap Name

Threshold

Value Unit

Description

Solution

Sev.

unexpected-clustersize

Expected Nodes

Node count

After a vCLUSTER startup time of 180 seconds has elapsed, the number of detected nodes should match cluster expectednodes configured. This trap is sent when the number of nodes has changed and is not the expected value.

Use the CLI to configure “cluster expected-nodes” to set the appropriate value.

Low

unexpectedshutdown

Event

Triggered

A Violin Memory Gateway has unexpectedly shutdown. This may happen during a software upgrade when the upgrade did not cleanly apply and the system reverted back to the previous version prior to upgrade. A callhome event is generated.


High

user-login

Event

Triggered

A login to a Violin Memory Gateway via either the CLI or Web UI has occurred.

Verify the user id has not been compromised.

Low

user-logout

Event

Triggered

A logout from a Violin Memory Gateway via either the CLI or Web UI has occurred.

Check that the appropriate settings for auto-logout are configured for the user. There are separate settings for the CLI vs. Web UI.

Low

vimm-state-change

State Change State

One or more VIMMs have changed state on a Violin Array. A comma separated list of VIMMs are provided for each state: admin_down, booting, active, spare, failed, present, or alarmed. The set value for each VIMM list / state change is “true” for set or “false” for clear.

Check the Violin Array for alarms to determine severity of state change.

High / Low

535-0008-00 Rev 10


447


Trap Name

vimm-temperaturehigh

Threshold

Value Unit

Description

Solution

Sev.

80

Celsius

VIMM temperature has exceeded normal operating range

Check airflow and operating environment.

High

vimm-temperature-ok 75

Celsius

VIMM temperature has dropped into normal operating range

Nothing.

Low

vmg-boot-raidchange

Active, Degraded, Resyncing, Unknown

There was a state change in the book disk. The states are as follows:

Degraded and Unknown status mean that the RAID disk is likely down and needs to be replaced.

High

NA

•

Active is a normal state.

•

Resyncing means the RAID group is rebuilding, which is also normal.

•

Degraded means the RAID disk is down.

•

Unknown means the RAID disk is unavailable and could be down.

vmg-fan-change

NA

Off, Absent, Low, Medium, High,

One or more fans have changed state on a Violin Memory Gateway.

An absent status could mean a failed fan. Replace the fan as soon as possible.

High

vmg-psu-change

NA

On, Off, Absent

Power supply status changed. There are two power supply units and both should be on.

If the status changes to off or absent, investigate the issue immediately.

High

448


535-0008-00 Rev 10

Configuring and Testing Traps

Configuring and Testing Traps Use the snmp command to configure traps, which can be sent to multiple hosts: violin-acma [violin: master] > enable violin-acma [violin: master] # conf terminal violin-acma [violin: master] (config) # snmp ? community contact in MIB-II enable host listen restrictions location in MIB-II port traps user

Add a new SNMP community Set a value for the syscontact variable Enable SNMP-related functionality Configure hosts to which to send SNMP traps Configure SNMP server interface access Set a value for the syslocation variable Set the UDP port for the SNMP agent Configure trap-related settings Configure SNMP access on a per-user basis

Enabling Traps Use the snmp-server traps event command to enable traps: violin [violin: master] (config) # snmp-server traps event cpu-util-high *violin [violin: master] (config) # wr mem█

Testing Traps Use the snmp-server traps send-test command to test traps: violin-acma [violin: master] (config) # snmp-server traps ? community Set the default community for traps sent to hosts which do not have a custom community string set event Specify which events will be sent as traps send-test Send a test trap violin-acma [violin: master] (config) # snmp-server traps send-test█ This will send a test trap to all configured trap sinks. This will be the 'testTrap' notification from the TMS-MIB. This trap is only ever sent on request from the user; it is not triggered automatically. This trap is not available for enable or disable through configuration; it is always enabled, meaning it will always be sent when requested by the user.

535-0008-00 Rev 10


449


Spare VIMMs To query the spare VIMMs in the system: [user@lab-2 mibs]$ snmpwalk -Ou -M +. -m +./VIOLIN-MEMORY-MIB.txt -v 2c -c public lab-ib-srv1 violin-memory | grep spare | grep "Gauge32: 1" enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.spa re."lab-system-098"."vimm30" = Gauge32: 1 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.spa re."lab-system-098"."vimm31" = Gauge32: 1 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.spa re."lab-system-098"."vimm32" = Gauge32: 1 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.spa re."lab-system-098"."vimm52" = Gauge32: 1█

Failed VIMMs To query the failed VIMMs in the system: snmpwalk -Ou -M +. -m +./VIOLIN-MEMORY-MIB.txt -v 2c -c public lab-ib-srv1 violinmemory | grep fail | grep "Gauge32: 1" [user@lab-2 mibs]$ (none)█

PSU States To query the state of the power supplies in the system: snmpwalk -Ou -M +. -m +./VIOLIN-MEMORY-MIB.txt -v 2c -c public lab-ib-srv1 violinmemory | grep -i psu enterprises.violinmemory.products.memoryGateway.appliance.media.chassisSystemArrayTable.chassisSy stemArrayEntry.chassisSystemPowerPSUA."lab-fender-098" = STRING: "ON" enterprises.violinmemory.products.memoryGateway.appliance.media.chassisSystemArrayTable.chassisSy stemArrayEntry.chassisSystemPowerPSUB."lab-fender-098" = STRING: "ON" [user@lab-ib-cn2 mibs]$█

450


535-0008-00 Rev 10


Temperatures: per VIMM and Chassis To query the temperatures (reported in degrees celsius) of the chassis and the VIMMs: [user@lab-ib-cn2 mibs]$ snmpwalk -Ou -M +. -m +./VIOLIN-MEMORY-MIB.txt -v 2c -c public lab-ib-srv1 violin-memory | grep -i temp enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm00" = INTEGER: 39 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm01" = INTEGER: 43 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm02" = INTEGER: 43 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm03" = INTEGER: 45 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm04" = INTEGER: 42 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm05" = INTEGER: 41 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm06" = INTEGER: 39 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm07" = INTEGER: 38 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm08" = INTEGER: 37 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm09" = INTEGER: 38 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm10" = INTEGER: 40 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm11" = INTEGER: 39 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm12" = INTEGER: 40 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm13" = INTEGER: 39 enterprises.violinmemory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.tem p."lab-system-098"."vimm14" = INTEGER: 39

535-0008-00 Rev 10


451


Performance Stats To query performance statistics for the system: [test@host1 mibs]$ snmpwalk -Ou -M +. -m +./VIOLIN-MEMORY-MIB.txt -v 2c -c public lab-vs3-n2 violin-memory | grep ata-VIOLIN enterprises.violinmemory.products.memoryGateway.appliance.media.mediaDevTable.mediaDevEntry.media DevIdx."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: ataVIOLIN_MEMORY_ARRAY_2110CR00000304 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaDevTable.mediaDevEntry.devId ."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: ataVIOLIN_MEMORY_ARRAY_2110CR00000304 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaDevTable.mediaDevEntry.fwVer sion."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: 3.7.2 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaDevTable.mediaDevEntry.g6Mod el."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: Violin Memory Array enterprises.violinmemory.products.memoryGateway.appliance.media.mediaBlockTable.mediaBlockEntry.d evId."2110CR00000304" = STRING: ata-VIOLIN_MEMORY_ARRAY_2110CR00000304 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaBlockTable.mediaBlockEntry.d evPath."2110CR00000304" = STRING: /dev/disk/by-id/ataVIOLIN_MEMORY_ARRAY_2110CR00000304-part4 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaStatsDevTable.mediaStatsDevE ntry.mediaStatsDevIdx."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: ataVIOLIN_MEMORY_ARRAY_2110CR00000304 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaStatsDevTable.mediaStatsDevE ntry.kernelOpsInProgress."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: 0 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaStatsDevTable.mediaStatsDevE ntry.kernelOpsTime."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: 47374 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaStatsDevTable.mediaStatsDevE ntry.kernelOpsTimeWeighted."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: 79475 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaStatsDevTable.mediaStatsDevE ntry.kernelReadCompleted."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: 1157474 enterprises.violinmemory.products.memoryGateway.appliance.media.mediaStatsDevTable.mediaStatsDevE ntry.kernelReadKbytes."ata-VIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: 11750483

452


535-0008-00 Rev 10


Fibre Channel (vSHARE) Information To query Fibre Channel information for a vSHARE configuration: enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.ww n.1."hba-a1" = STRING: wwn.21:00:00:24:ff:26:6a:8c enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.ww n.1."hba-a2" = STRING: wwn.21:00:00:24:ff:26:6a:8d enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.ww n.5."hba-a1" = STRING: wwn.21:00:00:1b:32:9f:d3:d1 enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.ww n.5."hba-a2" = STRING: wwn.21:01:00:1b:32:bf:d3:d1 enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.en able.1."hba-a1" = INTEGER: true(1) enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.en able.1."hba-a2" = INTEGER: true(1) enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.en able.5."hba-a1" = INTEGER: true(1) enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.en able.5."hba-a2" = INTEGER: true(1) enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.sp eed.1."hba-a1" = STRING: 8 Gbit enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.sp eed.1."hba-a2" = STRING: Unknown enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.sp eed.5."hba-a1" = STRING: 8 Gbit enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.sp eed.5."hba-a2" = STRING: Unknown enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.su pportedSpeeds.1."hba-a1" = STRING: 1 Gbit, 2 Gbit, 4 Gbit, 8 Gbit enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.su pportedSpeeds.1."hba-a2" = STRING: 1 Gbit, 2 Gbit, 4 Gbit, 8 Gbit enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.su pportedSpeeds.5."hba-a1" = STRING: 1 Gbit, 2 Gbit, 4 Gbit, 8 Gbit enterprises.violinmemory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.su pportedSpeeds.5."hba-a2" = STRING: 1 Gbit, 2 Gbit, 4 Gbit, 8 Gbit

535-0008-00 Rev 10


453


Fibre Channel Performance Statistics (vSHARE) Example The following is an example of querying Fibre Channel performance statistics for a vSHARE configuration. $ snmpwalk -c public -v 2c 10.1.10.137 VIOLIN-MEMORYMIB::statsTargetFcTable.statsTargetFcEntry -m VIOLIN-MEMORY-MIB.txt VIOLIN-MEMORY-MIB::statsTargetFcIdx."hba-a1" = STRING: hba-a1 VIOLIN-MEMORY-MIB::statsTargetFcIdx."hba-a2" = STRING: hba-a2 VIOLIN-MEMORY-MIB::txFrames."hba-a1" = STRING: 1222433 VIOLIN-MEMORY-MIB::txFrames."hba-a2" = STRING: 1222436 VIOLIN-MEMORY-MIB::rxFrames."hba-a1" = STRING: 611257 VIOLIN-MEMORY-MIB::rxFrames."hba-a2" = STRING: 611260 VIOLIN-MEMORY-MIB::dumpedFrames."hba-a1" = STRING: 0 VIOLIN-MEMORY-MIB::dumpedFrames."hba-a2" = STRING: 0 VIOLIN-MEMORY-MIB::nosCount."hba-a1" = STRING: 0 VIOLIN-MEMORY-MIB::nosCount."hba-a2" = STRING: 0 VIOLIN-MEMORY-MIB::linkFailureCount."hba-a1" = STRING: 116 VIOLIN-MEMORY-MIB::linkFailureCount."hba-a2" = STRING: 61 VIOLIN-MEMORY-MIB::lossOfSyncCount."hba-a1" = STRING: 0 VIOLIN-MEMORY-MIB::lossOfSyncCount."hba-a2" = STRING: 0 VIOLIN-MEMORY-MIB::lossOfSignalCount."hba-a1" = STRING: 0 VIOLIN-MEMORY-MIB::lossOfSignalCount."hba-a2" = STRING: 0 VIOLIN-MEMORY-MIB::invalidTxWordCount."hba-a1" = STRING: 0 VIOLIN-MEMORY-MIB::invalidTxWordCount."hba-a2" = STRING: 0 VIOLIN-MEMORY-MIB::invalidCrcCount."hba-a1" = STRING: 0 VIOLIN-MEMORY-MIB::invalidCrcCount."hba-a2" = STRING: 0 VIOLIN-MEMORY-MIB::primSeqProtocolErrCount."hba-a1" = STRING: 0 VIOLIN-MEMORY-MIB::primSeqProtocolErrCount."hba-a2" = STRING: 0█

New Trap MIB Objects Event: unexpected-cluster-size: The number of nodes in the cluster is unexpected. vmemClusterUnexpectedSize NOTIFICATION-TYPE OBJECTS { clusterConfigId, clusterConfigServiceDescr, clusterConfigExpectedNodes, clusterStateNumNodes, expectedNodesTypeUint16, currentNodesTypeUint16 } STATUS current DESCRIPTION "vmemClusterUnexpectedSize" ::= { vmemNotifTrapsPrefix 1 }

454


535-0008-00 Rev 10

New Trap MIB Objects

Event: unexpected-cluster-join: A node has unexpectedly joined the cluster. vmemClusterUnexpectedJoin NOTIFICATION-TYPE OBJECTS { clusterGlobalIdx, hostid, hostname, primaryAddr, clusterAddr, clusterStateNumNodes, clusterConfigId, clusterConfigServiceDescr, clusterConfigExpectedNodes, hostnameTypeString, expectedNodesTypeUint16, currentNodesTypeUint16 } STATUS current DESCRIPTION "vmemClusterUnexpectedJoin" ::= { vmemNotifTrapsPrefix 2 } Event: unexpected-cluster-leave: A node has unexpectedly left the cluster. vmemClusterUnexpectedLeave NOTIFICATION-TYPE OBJECTS { clusterGlobalIdx, hostid, hostname, primaryAddr, clusterAddr, clusterStateNumNodes, clusterConfigId, clusterConfigServiceDescr, clusterConfigExpectedNodes, hostnameTypeString, expectedNodesTypeUint16, currentNodesTypeUint16 } STATUS current DESCRIPTION "vmemClusterUnexpectedLeave" ::= { vmemNotifTrapsPrefix 3 }

535-0008-00 Rev 10


455


Event: media-device-health-warn: A media device has a health status warning. vmemHealthWarning NOTIFICATION-TYPE OBJECTS { id, location, type, vmemModel, vmemSerialNo, output } STATUS current DESCRIPTION "vmemHealthWarning" ::= { vmemNotifTrapsPrefix 11 } Event: media-device-lifetime-warn: A media device has a low estimated life remaining. vmemLifeRemain NOTIFICATION-TYPE OBJECTS { id, location, vmemModel, vmemSerialNo, vmemRevision, lifeRemain } STATUS current DESCRIPTION "vmemLifeRemain" ::= { vmemNotifTrapsPrefix 12 }

456


535-0008-00 Rev 10


Event: media-device-state-change: Unexpected change in media device state. vmemStateChange NOTIFICATION-TYPE OBJECTS { id, oldTypeString, newTypeString, devPath, blockSz, numBlocks, offset, vmemModel, devId, expected } STATUS current DESCRIPTION "vmemStateChange" ::= { vmemNotifTrapsPrefix 13 } Event: media-device-unknown-type: Detected a media device of unknown type. vmemUnknownDeviceType NOTIFICATION-TYPE OBJECTS { id, location, mfrModel, mfrSerialNo, size } STATUS current DESCRIPTION "vmemUnknownDeviceType" ::= { vmemNotifTrapsPrefix 14 } Event: media-system-swap-state: Detected a change in system swap state. vmemSystemSwapState NOTIFICATION-TYPE OBJECTS { active } STATUS current DESCRIPTION "vmemSystemSwapState" ::= { vmemNotifTrapsPrefix 15 }

535-0008-00 Rev 10


457


Event: vimm-temperature-high: High VIMM temperature detected. vmemVimmTempRisingError NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemVimmTempRisingError" ::= { vmemNotifTrapsPrefix 16 } Event: vimm-temperature-ok: VIMM temperature returned to normal levels. vmemVimmTempRisingClear NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemVimmTempRisingClear" ::= { vmemNotifTrapsPrefix 17 } Event: chassis-temperature-high: High chassis temperature detected. vmemChassisTempRisingError NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemChassisTempRisingError" ::= { vmemNotifTrapsPrefix 20 }

458


535-0008-00 Rev 10


Event: chassis-temperature-ok: Chassis temperature returned to normal levels. vmemChassisTempRisingClear NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemChassisTempRisingClear" ::= { vmemNotifTrapsPrefix 21 } Event: lid-ajar-time-rising-error: Chassis lid ajar time has exceeded a configured threshold. vmemLidAjarTimeRisingError NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemLidAjarTimeRisingError" ::= { vmemNotifTrapsPrefix 24 } Event: lid-ajar-time-rising-clear: Chassis lid ajar time condition is clear. vmemLidAjarTimeRisingClear NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemLidAjarTimeRisingClear" ::= { vmemNotifTrapsPrefix 25 } Event: array-pcie-link-up: Array PCIE link up. vmemArrayPcieLinkUp NOTIFICATION-TYPE OBJECTS { id, container } STATUS current DESCRIPTION "vmemArrayPcieLinkUp" ::= { vmemNotifTrapsPrefix 31 }

535-0008-00 Rev 10


459


Event: array-pcie-link-down: Array PCIE link down. vmemArrayPcieLinkDown NOTIFICATION-TYPE OBJECTS { id, container } STATUS current DESCRIPTION "vmemArrayPcieLinkDown" ::= { vmemNotifTrapsPrefix 32 } Event: array-data-plane-ready: Array data plane state changed. vmemArrayDataPlaneReady NOTIFICATION-TYPE OBJECTS { id, container, oldTypeUint8, newTypeUint8 } STATUS current DESCRIPTION "vmemArrayDataPlaneReady" ::= { vmemNotifTrapsPrefix 33 } Event: array-raid-rebuild: Array RAID rebuild state changed. vmemArrayRaidRebuild NOTIFICATION-TYPE OBJECTS { id, container, state } STATUS current DESCRIPTION "vmemArrayRaidRebuild" ::= { vmemNotifTrapsPrefix 34 }

460


535-0008-00 Rev 10


Event: vimm-state-change: Array VIMM state changed. vmemArrayVimmStateChange NOTIFICATION-TYPE OBJECTS { id, container, vimms, set, stateTypeString } STATUS current DESCRIPTION "vmemArrayVimmStateChange" ::= { vmemNotifTrapsPrefix 35 } Event: array-psu-state: Array PSU state changed. vmemArrayPsuState NOTIFICATION-TYPE OBJECTS { id, container, psuAChanged, psuAOld, psuANew, psuBChanged, psuBOld, psuBNew } STATUS current DESCRIPTION "vmemArrayPsuState" ::= { vmemNotifTrapsPrefix 36 }

535-0008-00 Rev 10


461


Event: array-led-change: Array LED state changed. vmemArrayLedChange NOTIFICATION-TYPE OBJECTS { id, container, alarmLedChanged, alarmLedOld, alarmLedNew, powerALedChanged, powerALedOld, powerALedNew, powerBLedChanged, powerBLedOld, powerBLedNew, statusLedChanged, statusLedOld, statusLedNew } STATUS current DESCRIPTION "vmemArrayLedChange" ::= { vmemNotifTrapsPrefix 37 } Event: fc-port-state-change: Fibre channel port state changed. vmemFcPortStateChange NOTIFICATION-TYPE OBJECTS { hostname, port, oldState, newState, speed } STATUS current DESCRIPTION "vmemFcPortStateChange" ::= { vmemNotifTrapsPrefix 38 }

462


535-0008-00 Rev 10


Event: array-fan-state-change: Array fan state changed. vmemArrayFanChange NOTIFICATION-TYPE OBJECTS { id, container, fan1Changed, fan1Old, fan1New, fan2Changed, fan2Old, fan2New, fan3Changed, fan3Old, fan3New, fan4Changed, fan4Old, fan4New, fan5Changed, fan5Old, fan5New, fan6Changed, fan6Old, fan6New } STATUS current DESCRIPTION "vmemArrayFanChange" ::= { vmemNotifTrapsPrefix 39 } Event: events-login: A user has logged in event. vmemEventsLogin NOTIFICATION-TYPE OBJECTS { username, timestamp, remoteAddr, peerId, clientDescr, sessionid } STATUS current DESCRIPTION "vmemEventsLogin" ::= { vmemNotifTrapsPrefix 40 }

535-0008-00 Rev 10


463


Event: events-logout: A user has logged out event. vmemEventsLogout NOTIFICATION-TYPE OBJECTS { username, timestamp, remoteAddr, peerId, clientDescr, sessionid } STATUS current DESCRIPTION "vmemEventsLogout" ::= { vmemNotifTrapsPrefix 41 }

464


535-0008-00 Rev 10

Optimizing Connectivity for Initiator Hosts

APPENDIX G

This appendix demonstrates ways in which you can optimize the connectivity of initiator (client-side) hosts with Violin Memory Arrays. The following topics are covered:

• •

Optimizing Connectivity for Windows Hosts on page 465 Optimizing Connectivity for AIX Hosts on page 466

Optimizing Connectivity for Windows Hosts Depending on the Windows operating system, client machines using vSHARE should use the Microsoft® Multipath I/O (MPIO) driver or the Violin Device Specific Module (DSM) driver for optimal storage array connectivity. Install the appropriate driver for your operating system.

Windows Server 2008 R2 and Later Install the Microsoft MPIO driver. MPIO is not installed by default. Use the following link for installation instructions. To install and configure MPIO for Windows Server 2008 R2, do the following: 1. Click the following link and follow the installation instructions.

http://technet.microsoft.com/en-us/library/ee619752(WS.10).aspx 2. To ensure a proper MPIO operation, complete the following steps: a. Install SP1 on Windows Server 2008 R2. b. Install HotFix KB2522766 (from the Microsoft Support site). c. Install HotFix KB2460971 (from the Microsoft Support site). d. Install HotFix KB2468345 (from the Microsoft Support site). e. Install HotFix KB2591462 (from the Microsoft Support site).

535-0008-00 Rev 10


465

Appendix G - Optimizing Connectivity for Initiator Hosts

3. Failover Clustering related: a. Install HotFix KB2531907 (from the Microsoft Support site).

Windows Server 2003 R2 This section demonstrates how to download and install the MPIO D4.5.0 driver, and then change the load balance policy setting. To download and install the MPIO D4.5.0 driver, do the following: 1. Go to the Violin Memory Web site (http://www.vmem.com). 2. Click

Support and then log in.

Note: You must be a registered Violin Memory customer to download this driver. If you are not registered, you are prompted to complete the registration process. 3. Click the

Software Downloads.

4. Click the

MPIO Driver DSM D4.5.0 link within Drivers and Utilities to download the software.

If you are using the Violin driver, the load balance policy is set to 1. For higher performance, this should be set to 2 (Round Robin). To change the load balance policy setting, complete the following steps: 1. Open the Registry Editor and navigate to:

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\v1010dsm\Parameters\Dsm LoadBalanceSettings\ 2. Modify DWORD: DsmLoadBalancePolicy and change the value to 2 for all discovered LUNs. 3. Restart the client machine.

Optimizing Connectivity for AIX Hosts The IBM (International Business Machines) initiator hosts running AIX operating systems support native multipathing capabilities through the Multipath I/O (MPIO) framework. It is recommended that AIX initiator hosts connected to Violin Memory Arrays have Violin drivers installed, along with AIX MPIO, to optimize connectivity. This section covers the following topics:

• • • • •

466

System Requirements on page 467 Installing the Violin Driver for AIX MPIO on page 467 Configuring an AIX Initiator Host for Optimum Connectivity on page 468 Discovering AIX Devices on page 469 Uninstalling the Violin Driver for AIX MPIO on page 469


535-0008-00 Rev 10

Optimizing Connectivity for AIX Hosts

System Requirements This section covers the requirements for AIX operating systems and Violin Memory Gateway software versions.

Supported AIX Operating Systems Violin drivers support the following versions of the AIX operating system:

• • •

AIX 6.1 (minimum supported TL: 6100-07-05-1228) AIX 7.1 (minimum supported TL: 7100-01-04-1216) VIO server version 2.1.3 with Fix Pack 23 (2.1.3.10-FP-23, or later)

Required AIX File Sets An AIX initiator host (Partition/DLPAR) requires the following file sets to effectively connect with the Violin Memory Arrays:

• • • •

devices.common.IBM.fc.rte devices.common.IBM.rte devices.fcp.disk.rte devices.fcp.tape.rte

Note: Refer to the AIX operating system documentation for instructions if any of the required file

sets listed above are not present on the AIX initiator host.

Required Violin Memory Gateway Software Violin Memory Arrays require Violin Memory Gateways running software version G5.5.1 or later to be interoperable with AIX clients.

Installing the Violin Driver for AIX MPIO This section demonstrates how to download and install the Violin driver for AIX MPIO.

Note: You can use smitty, the System Management Interface Tool, to install software on an AIX

system. Or, you can use the command suggested in the following procedure.

To download and install the Violin driver for AIX MPIO, do the following: 1. Go to the Violin Memory Support Site (http://www.violin-memory.com/support/). 2. Download the appropriate driver for AIX MPIO (devices.fcp.disk.vmem)—based on the version

of AIX running on your system—to a temporary location on the AIX initiator host, such as /var/tmp/violin.

535-0008-00 Rev 10


467



— Log in to the AIX initiator host as root user. — Log in to the VIOS partition as padmin, then enter a shell by running the following command. # oem_setup_env 4. Use smitty install, or the following command to install the Violin driver for AIX MPIO. # installp -acX -d/ devices.fcp.disk.vmem.rte

The following is an example of installing an AIX 6.1 initiator host. # install -acX -d/var/tmp/violin/6.1.03devices.fcp.disk.vmem.rte devices.fcp.disk.vmem.rte 5. Use the following command to verify that the Violin driver was successfully installed. # lslpp -l devices.fcp.disk.vmem.rte

Configuring an AIX Initiator Host for Optimum Connectivity This section covers the recommended configuration standards to obtain optimum connectivity with an AIX initiator host. 1. Configure the following settings for num_cmd_elems, as necessary:

•

Emulex LP10000 series FC HBA adapter: 1000

•


•


•

Adapter into VIO server configuration using NPIV capability: 2000

2. Ensure that the maximum transfer size (max_xfer_size) is set to 0x200000, using the

following commands: # chdev -l fcs0 -a max_xfer_size=0x200000 # chdev -l fcs0 -a num_cmd_elems=2000 3. Ensure the NACA bit is set for LUNs that are exported to AIX initiator hosts. For more

information, see the Violin 6000 Series Memory Array User’s Guide. 4. Configure the following settings for each fscsi driver object in the following way, to ensure non-

disruptive upgrades for Violin Memory Gateways. Note: The dyntrk parameter cannot be set to no for VIO/NPIV configurations. a. Set dyntrk to no using the following command: # chdev -l fscsi0 -a dyntrk=no -P b. Set fc_err_recov to delayed_fail using the following command: # chdev -l fscsi0 -a fc_error_recov=delayed_fail -P c. Reboot the AIX initiator host to enable the changes to the HBA settings.

468


535-0008-00 Rev 10

Optimizing Connectivity for AIX Hosts

Discovering AIX Devices This section demonstrates how to make exported Violin Memory LUNs visible to an AIX initiator host for the first time. To discover an AIX device, do the following: 1. Run the following command to initiate device discovery. # cfgmgr -v 2. To view Violin Memory LUNs run the following command. # lsdev -Cc disk

The Violin driver for AIX MPIO automatically configures the following attributes for optimum connectivity:

•

algorithm: round_robin

•

queue_depth: 255

•

reserve_policy: no_reserve

•

rw_timeout: 60

Uninstalling the Violin Driver for AIX MPIO This section demonstrates how to uninstall the Violin driver for AIX MPIO. Prerequisites:

•

Stop all applications that may be accessing LUNs on Violin Memory Arrays.

To uninstall the Violin driver for AIX MPIO, do the following: 1. Log in to the AIX initiator host as root user. 2. For LUNs controlled by a Logical Volume Manager (LVM), make the device unavailable (vary

off) before deleting LUNs from the Violin Memory Array. Use the following command. # varyoffvg 3. Ensure that the LUN entries are removed from the device tree, using the following command. # rmdev -dl 4. Use the following command to uninstall the Violin driver. # installp -ug devices.fcp.disk.vmem.rte 5. Use the following command to verify that the Violin driver has been successfully removed. E lslpp -l devices.fcp.disk.vmem.rte

535-0008-00 Rev 10


469


470


535-0008-00 Rev 10

System Specifications

This section contains specifications for the following:

• • • • •

535-0008-00 Rev 10

Memory Array System Specifications on page 472 Fibre Channel Adapter Card Specifications on page 473 InfiniBand Adapter Card Specifications on page 474 PCIe Host Adapter Card Specifications on page 474 Ethernet/iSCSI Adapter Card Specifications on page 475


471


Memory Array System Specifications The table below lists the Violin 6000 Series Memory Array system specifications:

Series

6600

Model

Violin Memory 6606

Flash Type

SLC

Raw Capacity (GiB/GB)

6TB/6.6TB

VIMM Count (Data + Hot Spare) 20 + 4

Violin Memory 6616

6200 Violin Memory 6212

Violin Memory 6232

MLC 16TB/17.6TB

12TB/13.2TB

32TB/35.2TB

60 + 4

20 + 4

60 + 4

Reliability/Resiliency

System Level Hardware Based vRAID 4 vRAID Controller Modules 2 Array Controller Modules 2 Memory Gateways Automated Failover & Failback

Connectivity Options

8x 4/8Gb/s Fibre Channel 8x 10GbE iSCSI 8x 40Gb/s QDR InfiniBand 4x PCIe Gen2 x8

Dimensions Height: Width: Depth: Cable Management:

5.25" (3RU/134 mm) 17.5” / 420 mm 27” / 700 mm 6” / 153 mm

Minimum Rear Clearance

7" (178 mm)

Weight

76lbs / 34.5kg

89lbs / 40.4kg

76lbs / 34.5kg

89lbs / 40.4kg

Power*

1110W

1679W

1344W

1709W

Cooling

3822 BTU/hr

5780 BTU/hr

4626 BTU/hr

5884 BTU/hr

Environmental Inlet Temp: Humidity: Altitude:

0°C to 35°C, to 7,500 feet 0°C to 30°C, 7,500 to 10,000 feet 5-95% (non condensing) 10,000 feet (Derate temperature)

Electrical Input Power Voltage: Operating Range: Current Rating:

200-240 VAC 180-264 VAC 15 A maximum

Interfaces

RJ-45 Serial Console (2) RJ-45 10/100/1000 Ethernet (2)

Management (CLI)

Serial, SSH, SNMP, Web/HTTP

Table G.1 Memory Array System Specifications *Nominal power measurements for Fibre Channel configuration.

472


535-0008-00 Rev 10

Fibre Channel Adapter Card Specifications

Fibre Channel Adapter Card Specifications The following table provides basic information for the QLogic QLE2562 Host Bus Adapter. See the QLogic Web site (http://www.qlogic.com) for the most complete and up-to-date information.

Fibre Channel Specifications Negotiation: IOPS: Class of service: Topology: Protocols: PCIe Express Interface Compliance:

8/4/2Gbps auto-negotiation 200,000 initiator and target IOPS per port 2 and 3 FC-AL, FC-AL2, point-to-point, switched fabric FCP-3-SCSI FC-Tape (FCP-2) PCIe Express Base Specification rev. 2.0 PCIe Express Card Electromechanical Specification rev. 2.0 PCIe Bus Power Management Interface Specification rev. 1.2 PCI Hot Plug Specification rev. 1.0

Physical and Electrical:

PCIe x8 physical connector StarPoint link training: - Maximum x4 lanes for Gen2 rate - Maximum x8 lanes for Gen1 rate

Connectivity Ports:

Dual 8Gbps Fibre Channel SFP+ with LC-style connector

HBA Specifications Airflow: Power consumption: Form factor: Temperature:

No airflow required 6.2 Watts (typical) Low-profile PCI Express card (6.6 in. x 2.54 in.) 0 to 55°C (operating) -40 to 70°C (non-operating) 10% to 90% (operating, non-condensing) 5% to 93% (non-operating, non-condensing) RoHS 6

Relative Humidity: RoHS compliance:

Table G.2 Fibre Channel HBA Specifications

The following table is a description of the LED states of the QLogic QLE2562 Host Bus Adapter. See the QLogic Web site (http://www.qlogic.com) for the most complete and up-to-date information.

Yellow (8Gbps)

Green (4Gbps)

Orange (2Gbps)

Off

Off

Off

Power off

On

On

On

Power on (before firmware initialization)

Flashing

Flashing

Flashing

Yellow, Green and Amber flashing alternatively

Description

Power on (after firmware initialization) Firmware error

Off

Off

On/Flashing

Online, 2Gbps link/I/O activity

Off

On/Flashing

Off


On/Flashing

Off

Off


Table G.3 LED Descriptions

535-0008-00 Rev 10


473


InfiniBand Adapter Card Specifications The following table provides basic information for the Mellanox ConnectX®-2 VPI Single and Dual Port QSFP InfiniBand and Ethernet Adapter Card. See the Mellanox Web site (http://www.mellanox.com) for the most complete and up-to-date information.

QSFP 40Gb/s Connector: Protocol Support InfiniBand:

InfiniBand (Copper and optical) Max power per port 2.0 W. Cable power budget class 2. IBTA v1.2.1, Auto-Negotiation (40Gb/s, 10Gb/s per lane), (20Gb/s, 5Gb/s per lane) or (10Gb/s, 2.5Gb/s per lane)

QoS:

8 InfiniBand Virtual Lanes for each port

Data rate:

QDR: InfiniBand

Connectivity:

- Interoperable with IB or 10GigE switches - microGiGaCN or QSFP connectors - Passive Copper cables (Direct Attach) - External optical media adapter and active cable support - QSFP to SFP+ connectivity through QSA

Table G.4 InfiniBand Adapter Card Specifications

PCIe Host Adapter Card Specifications The following table provides basic information for the Dolphin Express IXH610 Adapter. See the Dolphin Web site (http://www.dolphinics.com) for the most complete and up-to-date information.

Link Speeds:

40 Gbits/s

Application Performance:

0.81 microsecond latency (application to application)

Active Components:

IDT PES24NT6AG2 x8 Gen 2 PCI Express Chip

Topologies:

Point to point, Switched

Cable Connections:

One x8 Standard PCI Express Copper Cables

Power Consumption:

7 watts

Mechanical Dimensions:

PCI Express Card Electromechanical Specification 2.0

Operating Environment:

Operating Temperature: 0°C-55°C Relative Humidity: 5%-95% non-condensing

Table G.5 PCI Express Gen2 Host Adapter Specifications

474


535-0008-00 Rev 10

Ethernet/iSCSI Adapter Card Specifications

Ethernet/iSCSI Adapter Card Specifications The following table provides basic information for the Intel Ethernet X520 Adapter. See the Intel Web site (http://www.intel.com) for the most complete and up-to-date information.

Data rate supported per port:

Optical: 1 GbE/10 GbE Direct Attach: 10 GbE

Bus type:

PCI Express 2.0 (5 GT/s)

Bus width:

4-lane PCI Express and 8-lane PCI Express

Interrupt levels:

INTA, MSI, MSI-X

Operating temperature:

0°C-55°C (32°F-131°F)

Connectors:

One or two LC fiber-optic connectors SFP+ direct attach cables

Physical dimensions:

Low-profile PCI Express: 5.73 inches long (without PCI bracket

Table G.6 iSCSI Adapter Specifications

535-0008-00 Rev 10


475


476


535-0008-00 Rev 10

Index

Numerics 4kB alignment See partitions

A AC Power Failure, Shutting Down alarm 373 active configuration file 230, 238 backed up 231 admin role 159 access privileges 34, 119, 159 commands in Config mode 225 Admin user 34 admin username password 158 AIX configure initiator host 468 discover devices 469 file sets, required 467 hosts 466 MPIO 466 supported OS 467 uninstall Violin driver 469 Violin driver, installing 467 AIX compatibility 179 AIX compatibility, NACA 179 Alarm LED 370 alarms Alarm LED states 370 commands 371 non-service affecting alarms 377–385 service affecting alarms 372–377 ALUA attributes 257 definition 66 enable 66 535-0008-00 Rev 10

iSCSI-3 8 performance optimization 257 preferred port 257 settings, view 259 ARP configuration 319 Array Controller about 19 location 12 Array status screen 148 asterisk before CLI prompt 230, 238 authenticated sign-on 248 authentication bidirectional CHAP 248 CHAP 248 iSCSI CHAP 248 local 299 SMTP configuration 130 unidirectional CHAP 248

B bidirectional authentication 248 block size, of LUNs 63, 191, 334 block storage See vSHARE block storage media devices enabling or disabling 207–209 initializing Memory Arrays 180–181 read and write statistics 210 block zeroing, VAAI 212 Board Sensor Unreadable alarm 378 board status viewing in Web interface 149 Board Status page 49, 50 boot images 114 Booting Suspended, Fans Missing alarm 372


477

C cache media devices read and write statistics 210 call home auto support 128 fault reporting 128 notification 128 settings 128 call home settings 168 caution icons 3, 217 CHAP authenticated sign-on 248 bidirectional 248 bidirectional, configure 250 requirements 249 secret, establish 249 secret, remove 251 set up iSCSI 248 show configurations 252, 328 types 248 unidirectional 248 unidirectional, configure 250 chassis front view 13, 15–16 mounting on equipment rack 11 overview of 11 rear view 14 serviceability of 11 temperature alarms non-service affecting 378 service-affecting 376 top view of modules 12 CLI ARP configuration 319 boot commands 309 command history 291 configuration commands 306 connecting to 227 create user account 299 diagnostic commands 303 display settings 291 hidden commands 290 logging in 227–229 LUN statistics 216, 337 options 289 paging output 290 reboot 290 reload 290 See command line interface (CLI) Standard mode commands available 224 tcpdump 311 terminal type 291 user account commands 299 user inactivity 289 cluster Cluster Status page 48 management address, change 123, 163 Cluster Status page 48, 49 478

command array rebalance 242 command line interface (CLI) ambiguous command 222 asterisk for unsaved changes 230, 238 command completion finding options with ? 222 finding values with ? 224 tab completion 224 command descriptions 288–289 command list display with tabs 224 command modes 225 command prompts 226 INTERACTIVE commands 289 key to parameters 288 shorthand form 222 showing help 222 command modes Config mode 225 Enable mode 225 prompt and response conventions 226 Standard mode 225 user privileges 159 commands unified logging upload 219 Commit Changes button 230 Config mode 225 admin role 159 command prompt 226 entering 225 unsaved changes (asterisk) 226 configuration ARP 319 e-mail 129 network bond 315 RADIUS 328 reverting to saved configuration 232 save settings configuration file 230 file management 230, 238 saving settings 230, 238 showing 231 SMTP 130 TACAS+ 329 configuration file 230 active 230, 238 backed up 231 copying 232 deleting 232 file management 230 revert to saved configuration 232 save to a different file example 230 saving to a different file 230 show configuration files example 231 showing configuration 231 showing configuration files 231 active file 231


535-0008-00 Rev 10

switching to active file 230 configuration files changes unsaved to 226 configuration wizard running on demand 305 configuration write example 230 configure terminal Config mode 225 shorthand 222 connectivity optimize for AIX hosts 466 optimize for initiator hosts 465 optimize for Windows hosts 465 containers configuring 180 status 51 verifying activity and status 203 Controller FPGA Overheated, System Shut Down alarm 376

Controller FPGA Overheating alarm 378 customer support 5

D data management 253 best practices 255 scenarios 255 terminology 254 use cases 255 Data Plane Unavailable alarm 375 diagnostic tools 303 diskpart utility 426 DNS servers configuring 315 DNS settings 124, 164 domain name CLI parameter 288

E ECC errors 382, 384, 410 e-mail alert consolidation 217, 317 alerts 217 configuration 129 consolidation 217 customizing 218 E-mail Settings page 129, 162, 169 events 132 network settings 129 Enable mode command prompt 226 disable 225 monitor role 159 error messages ambiguous commands 222 535-0008-00 Rev 10

begin with % in CLI 226 global commands 225 Ethernet Interface 14, 472 Excessive ECC Errors on VIMM alarm 382

F Failed to Validate Configuration of VIMMs Raid Group alarm 382 failover protection 22 fan location 12 Fan Missing alarm 378 Fan Running Too Slowly alarm non-service affecting equipment alarm 379 fans 16 Fatal Flash Error, Hex-Value alarm 372 FATAL RAID Error alarm 374 fdisk utility 426 feature licenses 138, 170 Fibre Channel CLI commands 331 protocol 177 target ports 183 file management, configuration 306 firmware upgrading Memory Array 237 upgrading using CLI 233 flash alarms 373 non-service affecting 379–380 service-affecting 372 Flash CPL IRQs on VIMM alarm 379 Flash Format Failed on VIMM alarm 379 Flash Sequence Number Mismatch alarm 379 flash-based media media health 209 read and write statistics 210 show media 203 format changing format capacity 313, 320 default capacity 21 performance tuning 424 showing current format capacity 325 free space management 279 monitor 279 reclamation process 282 reclamation rules 283 set thresholds 283 thresholds 282 Front and Rear Panel page 54 front panel viewing in Web interface 56 full copy, VAAI 212


479

G global default gateway 123, 164

L

H help command completion tab completion 224 command-line help 222 question mark in CLI 222 showing in CLI 222 hidden commands, show 290 High Temperature on Board Sensor X X C alarm 378 host AIX initiator 466 initiator 465 Windows initiator 465 host, initiator 465 hostname CLI parameter 288 setting or clearing 315

I I/O Timeout From Host alarm 375 ID LED 16 remotely activating 56 igroups See initiator groups immediate upgrade, Memory Gateway 235 immediate upgrade, system 234 initiator groups 67 configuration of 187–188 create, delete 69 delete initiator 71 nested 72, 332 initiator hosts, connectivity 465 initiators create groups 69 delete from group 71 managing 69 Insufficient VIMMs alarm 376 INTERACTIVE commands 289 interfaces bonding of 315

ifname CLI parameter 288 physical ports 14 verifying activity and status 202 IP address CLI parameter 288 IP address CLI parameter 288 iSCSI authenticated sign-on 248 CHAP set up 248 target port configuration 184–187

480

targets, show 325 iSCSI-3, ALUA 8

large number abbreviations 226 LED identifying Memory Array 16 LEDs 15 Alarm LED, interpretation of states 370 libaio package 421, 430 libaio-devel package 421, 430 licenses, feature 138, 170 Linux drivers loading with modprobe 395 unloading 395 local authentication 299 logging automatic upload 297 CLI severity 296 delete oldest files 297 disable 296 formats 296 overrides 295 receive 296 remote server 295 remote upload 297 rotation 296 set severity 294 show commands 298 unified upload 219, 319 LUN group 254 LUN Management Container 59 Highlighted LUN, details 59 LUN statistics, CLI 216, 337 LUNs, information 59 LUN statistics, CLI 337 LUN Status page 51 LUNs 19 create, delete 62 creation of 63, 190–191, 333–334 export guidelines 194 exporting of 193–195, 336–337 LUN IDs 68, 186, 193, 195, 202, 337 Manage, Web Interface 59 Monitor LUNs page 110 monitoring, Web Interface 110 Read/Write Bandwidth 110, 111 Read/Write IOPs 110, 111 Read/Write Latency 110, 111 removing UDIDs 197 resizing 192 search and sort 60 specifying UDID 195 statistics, CLI 216


535-0008-00 Rev 10

M MAC address CLI parameter 288 Main Controller Does Not Support VIMMs alarm 377 Main Power Failure, System Shut Down alarm 373 Manage 221 LUNs 59 Manage Initiators page 69 Manage Targets page 75 management interfaces 27 RJ-45 Ethernet CLI 14 serial console port 14 serial console port CLI 227 mask netmask format 288 Master ACM 156 connecting to Violin Web interface 143 Master Gateway See master node master node 114 connect to Web Interface 31 statistics, viewing 48 media enable or disable all media 208 read and write statistics 210 media, health status 209 Memory Array about 8 alarms understanding 369 rebooting 239 shutting down modules 239 version information 157 Memory Arrays, initializing for block storage 180–181 Memory Gateway boot images 114 master node, Web Interface connection 31 Memory Gateways about 19 location 12 shutting down 236–237 memory plane 18 memory, show media 203 message bar Commit Changes button 230 modules powering off and on 240 shutting down 239 Monitor Monitor LUNs page 110, 111 monitor role 159 access privileges 34, 119, 159 commands in Enable mode 225 Monitor user 34 monitor username password 158 MTBF 18

535-0008-00 Rev 10

Multiple RAID Errors alarm 374 Multiple VIMMs Failed 377

N NACA 179 NACA bit support 179 netmask CLI parameter 288 network bond 315 Network Interfaces about 19 location 13 network interfaces See interfaces No PCIe Host Detected on Enabled Port alarm 373 No Spare Memory Modules alarm 382 nodes, cluster cluster node id CLI parameter 288 NTP modify settings 125

O OpenVMS 195 over provisioning 255, 256

P partitions 4kB alignment 425–427 creating 397–400 password admin 158 change 120, 160 monitor 158 PCIe switching 10 performance optimization 257 performance testing using Perf_Test 428–432 port CLI parameter 288 Port X Negotiated to 0 Lanes alarm 373 ports alarms 373, 380 POSIX threads 420 Power Controllers about 18 location 12 power supplies 16 alarms 373, 380 Power Supply Failure alarm 380 Power Supply Missing alarm 380 Power Supply Polling Failure alarm 380


481

preferred port, ALUA 257 Preparing Flash on VIMM alarm 379 Programming VIMM alarm 382 provision over 255 thin 254, 260 thresholds, alert, limit 103 usage thresholds 256 proxy settings, Web 311

R RADIUS configuration 328 RAID alarms 374, 380–381 automatic balancing 244 discovering unbalanced group 243 rebalance commands 242 rebalancing 242 scheduling group rebalance 245 showing rebalance settings 245 RAID Consistency Error alarm 374 RAID Group Faulted alarm 374 RAID Group Unprotected alarm 381 RAID Rebuild in Progress alarm 381 Raid Rebuild Of Group %D Completed with X Errors Starting at 4K Address Y alarm 381 RAID Rebuild of Group Completed with Errors (No Additional Information Available) alarm 381 RAID V2P Error alarm 374 rear panel viewing in Web interface 54 reboot 290 REST 27 REST API 29 revert to saved configuration example 232 roles specify for a user 119, 159 user privileges 119, 158 routing commands 316

S save configuration 230, 238 active configuration file 230, 238 configuration file 230, 238 configuration write 230 write memory 230 schedule modify schedule 276 schedule management 274 create schedule 274 delete schedule 275 show list 279 482

Scheduler Paused alarm 375 SCSI sessions 186, 202 SCSI-3, ALUA 66 serial cable connecting 227 server, xinetd 311 show configuration CLI command options 231 show configuration files example 231 show configuration running 231 show hidden commands 290 show media 203 media health 209 options 203 show running-config 231 show stats media 210 show system alarms command 371 show system message command 371 sign-on, authenticated 248 SMTP, authentication configuration 130 snap group 254, 270 create 270 delete 271 export 273 modify 272 rename 273 rollback 272 show list 271 snapshots 263 create 264 delete 264 export 269 modify access to 269 protect 270 protected 254 rename 268 rollback 267 view information 265 SNMP 27 configure 134 events 136 management 133 settings, Web Interface 133 trap sinks 135 SNMP Traps 29 software downloading 233 upgrading 233 Memory Gateway 235 upgrading using CLI 233 SSH 14 staged upgrade, Memory Gateway 235 Standard mode 225 command prompt 226 unpriv role 159 statistics, media read and write 210 storage media


535-0008-00 Rev 10

read and write statistics 210 show media options 203 status 209 Syslog 27, 28 System Booting alarm 376 System Shutdown alarm 375, 377 System Status LED 15, 370 System View 48

control inactivity 289 edit account 120, 160 password initial value 158 privileges admin role 34, 119, 159, 225 monitor role 34, 119, 159, 225 unpriv role 225 privileges and roles 119, 158, 159 user management 119 user accounts 119, 299

T TACAS+ configuration 329 target ports configuring 182 Fibre Channel ports 183 iSCSI ports, configuration of 184–187 verifying port activity and status 201 targets manage 75 Manage Targets page 75 Monitor Targets page 111 monitoring, Web Interface 111 TCP port range of values 288 tcpdump, commands 311 Telnet 14 telnet command 224 thin provisioning 254, 260 thresholds, alert, reclaim 103 thresholds, free space 283 thresholds, usage 256 tools, diagnostic 303 trap sinks, SNMP 135

V VAAI support 212 vcounts command 406–409 veeprom command 409 VIMM about 9 VIMM Booting (X% Complete) alarm 383 VIMM Configuration Out of Spec alarm 383 VIMM Extremely Hot 384 VIMM Failed Excessive ECC Errors alarm 384 VIMM Failed alarm 384 VIMM Failed alarms 383 VIMM Flash Format In Progress alarm 377 VIMM Hot alarm 384 VIMM in Maintenance State alarm 385 VIMM Incompatible Formatted Capacity alarm 385 VIMM Isolated alarm 385 VIMM Shutdown alarm 384 VIMM status 49 VIMM Status LED 15, 370 VIMM Temperature Unreadable, VIMM Shut Down alarm 385

U Unable to Poll VIMM Temperatures 375 Unformatted System alarm 373 unidirectional authentication 248 Unified Logging Upload 219, 319 unpriv role commands in Standard mode 225 Unsupported VIMM Configuration alarm 382 upgrading Memory Array firmware 237 Memory Gateway software 235 URL CLI parameter 288 usage thresholds 256 usage thresholds, set 283 USB LEDs 15 used space thresholds 103 user change password 120, 160 535-0008-00 Rev 10

VIMM Threshold Exceeded alarm 385 VIMM Unformatted alarm 385 VIMMs 20 alarms 376–377, 381–385 architecture 20 configurations 22 flash-based 21 location 13 slot locations 23, 24, 26 spare locations 23 spares 23 types of 21 vincident command 410 vinfo command 410–412 Violin 6000 architecture 9 components 17–26 models 10 performance of 21 platform 11 platform components 17


483

system specifications 472 Violin driver AIX MPIO, install 467 AIX MPIO, uninstall 469 Violin Linux driver installing 388–392 loading manually 394–395 overview 387 uninstalling 393 Violin Utilities installing 396, 403 running 403 Violin Web Interface layout, overview 32 supported web browsers 33 task map 39 VMware VAAI support 212 vpartial command 413–414 vpartial utility 4kB alignment, assessment of 425 vRAID about 10 vRAID Controllers about 19 location 13 vring utility 4kB alignment, assessment of 425–426 command 415 vSHARE 19 architecture 176 configuration 177–178 vstat command 416–418 VXM 19 VXM technology 9 vzero command 419

W warning icons 3 Web Admin page 126, 167 Web Interface Administration 113, 154 Board Status page 49 Call Home Settings 128, 168 Cluster Status page 48

484

Commit Changes button 230 container status 51 DNS Settings page 124, 164 E-mail Events 132 E-mail Settings 129, 162, 169 feature licenses 138, 170 LUN Status page 51 LUN status page 51 LUNs, monitor 111 Manage LUNs 59 Monitor LUNs page 110 Monitor Targets page 111 Monitoring Overview page 106 NTP Settings page 125 RAID group 50 SNMP Events 136 SNMP Settings 133 supported Web browsers 145 targets, monitor 111 task map 39 Web Admin page 126, 167 Web interface configuring 146 Front and Rear Panel page 54 printing screens 146 Web proxy, settings 311 Windows initiator host 465 optimized connectivity 465 Server 2003 R2 466 Server 2008 R2 465 write memory example 230 write terminal 231

X xinetd server, settings 311 XTree interface accessing 173

Z zeroing-out data 191, 334


535-0008-00 Rev 10

Violin 6000 Series Memory Array User's Guide - Diamond Point ...

Violin 6000 Series Memory Array User's Guide - Diamond Point ...

Suggest Documents

Violin 6000 Series Memory Array Installation Guide - Diamond Point ...

Violin Memory 6000 Flash Memory Array Datasheet

GeoExplorer 6000 Series User Guide

GeoExplorer 6000 Series User Guide

Violin Memory - Violin Systems

Violin Memory - Violin Systems

Demartek Evaluation of Violin Memory 7300 All Flash Array for ...

Demartek Evaluation of Violin Memory 7300 All Flash Array for ...

ADAM 6000 Series Manual

B Series Users Guide - Bombjack.org

TSP 6000 Series

tp-6000 series - pdfstream.manualsonline.com

Nominee:Violin Memory Nomination title: 6000 Series ... - DCS Awards

HP Officejet 6000 (E609) Printer Series User Guide

Series 6000 digital phone user guide - UNH Information Technology

Canon IR 5000/6000 series

HP Officejet 6000 (E609) Printer Series User Guide

MPro 200 Series Users Guide - JBL Professional

HP Officejet 6000 (E609) Printer Series User Guide

MPro 400 Series Users Guide - JBL Professional

Recurrent Memory Array Structures

Recurrent Memory Array Structures

Demartek Evaluation of Violin Memory 7300 All Flash ... - Violin Systems

Demartek Evaluation of Violin Memory 7300 All Flash ... - Violin Systems