NetWorker V7.3, Disaster Recovery Guide

NetWorker V7.3, Disaster Recovery Guide

NetWorker V7.3

Disaster Recovery Guide

Edition August 2006

CommentsSuggestionsCorrections

The User Documentation Department would like to know your opinion on this manual. Your feedback helps us to optimize our documentation to suit your individual needs.

Fax forms for sending us your comments are included at the back of the manual.

There you will also find the addresses of the relevant User

Documentation Department.

cognitas. Gesellschaft für Technik-Dokumentation mbH www.cognitas.de

Copyright and Trademarks

Base of NetWorker V7.3 Disaster Recovery Guide / this manual:

EMC Legato NetWorker

®

, Release 7.3, Multiplatform Version, Disaster Recovery Guide, Copyright © 1990 - 2005 by EMC Corporation.

Copyright © Fujitsu Siemens Computers GmbH 2006.

All rights reserved.

Delivery subject to availability; right of technical modifications reserved.

All hardware and software names used are trademarks of their respective manufacturers.

Product Number: 0604600

Contents

Preface ......................................................................................................... 11

.

Audience.......................................................................................................................... 11

.

Organization .................................................................................................................... 11

.

Related Documentation ................................................................................................... 12

.

Conventions Used in This Guide ..................................................................................... 13

.

Where to Get Help ........................................................................................................... 14

.

Technical Support ..................................................................................................... 14

.

Customer Service...................................................................................................... 14

.

Customer Feedback ........................................................................................................ 14

.

Chapter 1: Introduction............................................................................... 15

.

Supported Operating Systems ........................................................................................ 15

.

Definition of a Disaster .................................................................................................... 16

.

Chapter 2: Preparing for a Disaster........................................................... 17

.

Backing Up Data.............................................................................................................. 17

.

Gathering the Key Information......................................................................................... 18

.

Hardware Information................................................................................................ 18

.

Software Information ................................................................................................. 19

.

Chapter 3: UNIX Disaster Recovery........................................................... 21

.

Disaster Recovery Procedures ........................................................................................ 22

.

Replacing a Hard Drive ................................................................................................... 23

.

Prerequisites ............................................................................................................. 23

.

Replacing the Hard Drive .......................................................................................... 23

.

Completing the Recovery Process............................................................................ 24

.

Reinstalling and Configuring the UNIX Operating System .............................................. 24

.

NetWorker V7.3

Disaster Recovery Guide

3

Contents

Prerequisites.............................................................................................................. 24

.

How to Reinstall and Configure the UNIX Operating System.................................... 25

.

Completing the Recovery Process ............................................................................ 28

.

Recovering a NetWorker Server ...................................................................................... 28

.

Prerequisites.............................................................................................................. 28

.

How to Recover a NetWorker Server ........................................................................ 29

.

Recovering a NetWorker Storage Node........................................................................... 38

.

Prerequisites.............................................................................................................. 39

.

How to Recover a NetWorker Storage Node............................................................. 39

.

Recovering a NetWorker Client........................................................................................ 42

.

Prerequisites.............................................................................................................. 42

.

How to Recover a NetWorker Client.......................................................................... 43

.

Recovering the NetWorker Management Console Server ............................................... 45

.

How to Recover the Console Server Database in a Standalone Environment.......... 46

.

How to Recover the Console Server Database in a Cluster Environment ................ 47

.

Chapter 4: Windows Disaster Recovery ....................................................49

.

Disaster Recovery Summary of Procedures .................................................................... 50

.

Summary of a NetWorker Server Recovery .............................................................. 50

.

Summary of a NetWorker Client or Storage Node Recovery .................................... 53

.

Disaster Recovery of Multiple Computers........................................................................ 54

.

Disaster Recovery of Clustered Computers .............................................................. 54

.

Replacing a Hard Drive .................................................................................................... 55

.

Performing a Windows ASR Recovery ............................................................................ 55

.

Limitations of the Operating System’s Recovery CD-ROM ....................................... 56

.

ASR Recovery Requirements.................................................................................... 57

.

Task 1: Perform an ASR Recovery of the NetWorker Client Computer .................... 57

.

Task 2: Recover Components that Require Special Handling .................................. 59

.

Task 3: Verify the NetWorker Client Recovery .......................................................... 60

.

Repairing a Windows NT 4.0 Operating System Installation ........................................... 61

.

Windows NT 4.0 Recovery Requirements................................................................. 61

.

4

NetWorker V7.3

Disaster Recovery Guide

Contents

Creating an ERD from the REPAIRDISK Save Set .................................................. 61

.

How to Use the ERD to Repair the Windows NT 4.0 Installation.............................. 63

.

Recovering a Windows Installation.................................................................................. 64

.

Windows Recovery Requirements ............................................................................ 65

.

How to Recover the Windows Operating System ..................................................... 66

.

Recovering a NetWorker Server...................................................................................... 71

.

NetWorker Server Recovery Requirements.............................................................. 71

.

How to Recover a Windows NetWorker Server ........................................................ 71

.

Recovering a NetWorker Client or Storage Node............................................................ 87

.

Partial Recovery of a NetWorker Storage Node ....................................................... 88

.

Task 1: Satisfy the NetWorker Client or Storage Node Recovery Requirements ..... 88

.

Task 2: Reinstall the NetWorker Client or Storage Node Software........................... 89

.

Task 3: Recover the NetWorker Client or Storage Node Data.................................. 89

.

Task 4: Verify the NetWorker Client or Storage Node Recovery .............................. 92

.

Recovery of DHCP and WINS Databases....................................................................... 93

.

How to Recover a DHCP Database .......................................................................... 94

.

How to Recover a WINS Database........................................................................... 94

.

Recovering the NetWorker Management Console Server .............................................. 94

.

How to Recover the NetWorker Management Console Server Database ................ 95

.

Chapter 5: Mac OS X Disaster Recovery................................................... 97

.

Recovering a NetWorker Client ....................................................................................... 97

.

Prerequisites ............................................................................................................. 97

.

How to Recover a NetWorker Client ......................................................................... 98

.

Chapter 6: NetWare Disaster Recovery................................................... 101

.

Replacing a Hard Drive ................................................................................................. 101

.

Obtain the Hard Drive Information .......................................................................... 102

.

Replace the Hard Drive........................................................................................... 102

.

Completing the Recovery Process.......................................................................... 103

.

Recovering the NetWare Operating System ................................................................. 103

.

NetWorker V7.3

Disaster Recovery Guide

5

Contents

Prerequisites ........................................................................................................... 103

.

How to Recover the NetWare Operating System .................................................... 104

.

Completing the Recovery Process .......................................................................... 106

.

Recovering a NetWare NetWorker Server ..................................................................... 107

.

Prerequisites ........................................................................................................... 107

.

How to Recover a NetWare NetWorker Server ....................................................... 107

.

Recovering a Replicated NDS Partition ......................................................................... 114

.

How to Recover a NetWare 4.10 SYS Volume ....................................................... 115

.

How to Recover a NetWare 4.11, NetWare 5.0, or IntraNetWare Server

SYS Volume ............................................................................................................ 118

.

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0) ..... 123

.

Recovering One Node in a Cluster ................................................................................ 123

.

Task 1: Evict Node_B from the Cluster ................................................................... 124

.

Task 2: Reinstall Windows NT 4.0 and MSCS on Node_B ..................................... 124

.

Task 3: Reinstall the NetWorker Software............................................................... 125

.

Task 4: Recover the Data on Node_B ..................................................................... 125

.

Recovering a Cluster Shared Disk ................................................................................. 125

.

Task 1: Take the Dependent Resources Offline...................................................... 126

.

Task 2: Replace the Hard Drive .............................................................................. 126

.

Task 3: Recover the Data to the Shared Disk ......................................................... 127

.

Task 4: Configure the Cluster to Include the Shared Disk....................................... 127

.

Replacing the Quorum Disk and Recovering Its Data.................................................... 128

.

About the Quorum Resource ................................................................................... 128

.

How to Replace the Quorum Disk and Recover Its Data ........................................ 129

.

Recovering the MSCS Cluster Configuration................................................................. 129

.

Task 1: Recover the System State and Quorum Data ............................................ 130

.

Task 2: Replace the Quorum Data with the Recovered Quorum Data.................... 130

.

Task 3: Evict Node_B from the Cluster ................................................................... 131

.

Task 4: Reinstall the Cluster Software and Patches on Node_B ............................ 131

.

Performing a Cluster-Wide Recovery............................................................................. 131

.

6

NetWorker V7.3

Disaster Recovery Guide

NetWorker V7.3

Disaster Recovery Guide

Contents

Task 1: Reinstall and Configure the Software on Node_A...................................... 132

.

Task 2: Reinstall and Configure the Software on Node_B...................................... 133

.

Task 3: Reinstall and Configure NetWorker............................................................ 133

.

Task 4: Perform a Disaster Recovery ..................................................................... 134

.

Task 5: Check and Reassign the Drive Assignments ............................................. 134

.

Task 6: Recover the Data ....................................................................................... 135

.

Task 7: Check the Disk Signatures ......................................................................... 136

.

Task 8: Join Node_B to the Cluster ........................................................................ 137

.

Chapter 8: Microsoft Cluster Disaster Recovery

(Windows 2000 and 2003)......................................................................... 139

.

Prerequisites.................................................................................................................. 139

.

Recovering a Failed Quorum Disk................................................................................. 140

.

How to Recover a Failed Quorum Disk................................................................... 140

.

Recovering One Cluster Node....................................................................................... 141

.

How to Recover One Cluster Node......................................................................... 142

.

Recovering Multiple Cluster Nodes ............................................................................... 144

.

How to Recover Multiple Cluster Nodes ................................................................. 144

.

Troubleshooting Cluster Database Recovery................................................................ 147

.

Cluster Recovery Environment Variable Settings ................................................... 147

.

Chapter 9: VERITAS Cluster Server Disaster Recovery ........................ 149

.

Prerequisites.................................................................................................................. 149

.

Recovering a Single Cluster Node ................................................................................ 150

.

How to Recover a Single Cluster Node................................................................... 150

.

Recovering an Entire Cluster......................................................................................... 151

.

How to Recover an Entire Cluster........................................................................... 151

.

Chapter 10: Sun Cluster Disaster Recovery ........................................... 155

.

Prerequisites.................................................................................................................. 155

.

Recovering a Single Cluster Node ................................................................................ 155

.

7

Contents

How to Recover a Single Cluster Node ................................................................... 156

.

Recovering an Entire Cluster ......................................................................................... 156

.

How to Recover an Entire Cluster ........................................................................... 156

.

Chapter 11: HP TruCluster Disaster Recovery........................................ 159

.

Prerequisites .................................................................................................................. 159

.

Record Information .................................................................................................. 159

.

Back Up the Cluster................................................................................................. 160

.

Recovering One or More Nodes in a Cluster ................................................................. 160

.

How to Recover One or More Nodes in a Cluster ................................................... 161

.

Recovering an Entire Cluster ......................................................................................... 163

.

How to Recover an Entire Cluster ........................................................................... 163

.

Chapter 12: HP-UX Cluster Disaster Recovery ....................................... 175

.

Recovering a Single Cluster Node ................................................................................. 175

.

How to Recover a Single Cluster Node ................................................................... 175

.

Performing a Cluster-Wide Recovery............................................................................. 176

.

How to Perform a Cluster-Wide Recovery............................................................... 176

.

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery ................................................................. 179

.

Recovering a Single Cluster Node ................................................................................. 179

.

How to Recover a Single Cluster Node ................................................................... 179

.

Performing a Cluster-Wide Recovery............................................................................. 181

.

How to Perform a Cluster-Wide Recovery............................................................... 181

.

Renaming the Resource Directory ................................................................................. 186

.

Chapter 14: EMC FullTime AutoStart for Windows Disaster Recovery 189

.

Prerequisites .................................................................................................................. 189

.

Recovering a Single Cluster Node ................................................................................. 189

.

8

NetWorker V7.3

Disaster Recovery Guide

Contents

Chapter 15: HACMP for AIX Disaster Recovery ..................................... 191

.

Recovering a Single Node............................................................................................. 191

.

How to Recover a Single Node ............................................................................... 192

.

Recovering an Entire Cluster......................................................................................... 192

.

How to Perform a Cluster-Wide Recovery .............................................................. 192

.

Glossary ..................................................................................................... 197

.

Index ........................................................................................................... 217

.

NetWorker V7.3

Disaster Recovery Guide

9

Contents

10

NetWorker V7.3

Disaster Recovery Guide

Preface

This guide is part of the Fujitsu Siemens Computers GmbH NetWorker

®

V7.3 for Solaris™/SPARC and Linux software documentation set. It also comprises the EMC

Microsoft

2

NetWorker descriptions for other operating systems, such as

®

Windows

®

. For more information on available documentation, see

"Related Documentation" on page 12 .

Audience

The information in this guide is intended for the following audiences:

• System administrators who perform backup and recovery procedures, and maintain the safety of the data located over a network.

• Managers who want to learn how to implement a disaster recovery program.

• Users who are responsible for implementing disaster recovery plans and procedures.

Organization

This guide is organized as follows:

"Chapter 2: Preparing for a Disaster" on page 17

"Chapter 3: UNIX Disaster Recovery" on page 21

"Chapter 4: Windows Disaster Recovery" on page 49

"Chapter 5: Mac OS X Disaster Recovery" on page 97

"Chapter 6: NetWare Disaster Recovery" on page 101

"Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)" on page 123

NetWorker V7.3

Disaster Recovery Guide

11

Preface

"Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and

2003)" on page 139

"Chapter 9: VERITAS Cluster Server Disaster Recovery" on page 149

"Chapter 10: Sun Cluster Disaster Recovery" on page 155

"Chapter 11: HP TruCluster Disaster Recovery" on page 159

"Chapter 12: HP-UX Cluster Disaster Recovery" on page 175

"Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris

Disaster Recovery" on page 179

"Chapter 14: EMC FullTime AutoStart for Windows Disaster Recovery" on page 189

"Chapter 15: HACMP for AIX Disaster Recovery" on page 191

Related Documentation

The complete set of the Fujitsu Siemens Computers NetWorker documentation can be found at www.fujitsu-siemens.com/networker. Follow the link

NetWorker backup solution. Most of the documents are in Adobe

®

Portable

Document Format (PDF), and can be viewed by downloading and installing the Adobe (Acrobat) Reader as of V5.0. The Reader is available from Adobe at www.adobe.com

. To install and use the Reader on the platform of your choice, refer to the instructions at the Adobe web site.

For more information on Fujitsu Siemens Computers NetWorker software, refer to the Administrator’s Guide.

12

NetWorker V7.3

Disaster Recovery Guide

Preface

Conventions Used in This Guide

This document uses the following typographic conventions and symbols to make information easier to access and understand.

Convention Indicates Example boldface

italic in text

Names of line commands, daemons, options, programs, or scripts

The nsradmin command starts the command line version of the NetWorker

Administrator program.

Pathnames, filenames, computer names, new terms defined in the Glossary or within the chapter, or emphasized words

Displayed messages are also written to

/nsr/logs/daemon.log.

save -s

server_name

italic in command line

A variable that must be provided in the command line fixed-width

Examples and information displayed on the screen media waiting: recover waiting for

8mm 5GB tape volume name

fixed-width, boldface

Commands and options that must be entered exactly as shown

nsr_shutdown -a

Menu_Name>

Command

A path or an order to follow for making selections in the user interface

Volume>Change Mode>Appendable

Important:

Information that must be read and followed to ensure successful backup and recovery of data

Important:

Use the no_verify option with extreme caution.

NetWorker V7.3

Disaster Recovery Guide

13

Preface

Where to Get Help

Company, product, and technical information can be obtained as follows.

Technical Support

The Support section of the Fujitsu Siemens Computers web site at http://www.fujitsu-siemens.com/service/ provides links with contact information, software patches, technical documentation, and information about available support programs.

• If you have an active support agreement, you may contact your service provider.

• If you do not have an active support agreement, contact your Fujitsu

Siemens Computers sales representative to purchase a service contract or updates.

Customer Service

Technical Support Service Address

Latest news about NetWorker and software patches www.fujitsu-siemens.com/networker

NetWorker manual pages

NetWorker training http://manuals.fujitsu-siemens.com

[email protected]

Support phone

Support fax

+49-180-54040

+49-180-5336779

Customer Feedback

Please send any comments, suggestions and corrections on the NetWorker V7.3

Disaster Recovery Guide using the form at the end of this manual.

14

NetWorker V7.3

Disaster Recovery Guide

Chapter 1: Introduction

The NetWorker V7.3 Disaster Recovery Guide provides step-by-step instructions for recovering from a disaster on supported NetWorker V7.3 client and server platforms.

This chapter includes the following sections:

"Supported Operating Systems" on page 15

"Definition of a Disaster" on page 16

Supported Operating Systems

This guide provides disaster recovery information for NetWorker software on the following operating systems:

• Linux

• Mac OS X

• Microsoft Windows

• NetWare

• UNIX

NetWorker V7.3

Disaster Recovery Guide

15

1

Chapter 1: Introduction

Definition of a Disaster

For the purpose of this guide, a disaster is any situation in which the day-to-day access to data (for example, working files, software programs, or system files) is disrupted. A disaster can also damage network components, such as data, devices, hardware, media, and software.

A disaster can result from any of the following:

• Computer viruses that can corrupt data

• Hardware and software failures

• Infrastructure interruptions, inconsistencies, or loss of services, such as communication or network connections

The degree of loss during a disaster can range from one or more files to an entire computer system. The severity of the disaster determines the procedures necessary to recover data.

Note:

The term autochanger refers to a variety of backup devices: autoloader, carousel, datawheel, jukebox, library, and near-line storage.

16

NetWorker V7.3

Disaster Recovery Guide

Chapter 2: Preparing for a Disaster

This chapter includes the following sections on preparing for a disaster.

"Backing Up Data" on page 17

"Gathering the Key Information" on page 18

If you are viewing the online version of this guide, print a hard copy and store it in a safe location.

Backing Up Data

Back up important data on a scheduled basis. The more time and effort invested in incorporating, maintaining, and testing a backup solution, the better prepared you are in the event of a disaster.

Ensure that servers are backed up regularly as part of a backup group.

Otherwise, a

bootstrap

is not saved (backups performed by using the command line or the NetWorker User program do not save the NetWorker server bootstrap). You should also use a local backup device on the server to back up the server’s bootstrap.

Always perform a scheduled backup of the NetWorker server after upgrading to a new version of NetWorker software. This ensures that an upgraded version of the bootstrap is saved.

The bootstrap information can be printed from the savegrp.log file, which is located in the nsr directory. For instructions on how to configure the

NetWorker software to send bootstrap information directly to a printer or to a specified e-mail address, refer to the NetWorker Administrator’s Guide. If the bootstrap is backed up to a pool other than the preconfigured pools, save the name of the pool along with the bootstrap.

NetWorker V7.3

Disaster Recovery Guide

17

2

Chapter 2: Preparing for a Disaster

Maintain a copy of this information in a safe location for easy access by those assigned to perform disaster recovery. Consider maintaining a copy of this information in both an onsite and an offsite location.

Important:

Having the correct information on hand in case a disaster occurs is a key element in recovering from a disaster as quickly as possible.

Gathering the Key Information

Maintain accurate records for each hardware, software, network, device, and media component.

Hardware Information

Maintain the following up-to-date information regarding computer hardware:

• File-system configuration

• Fully qualified domain names, IP addresses, and hostnames

• For Domain Name System (DNS) clients, maintain the DNS host’s

Internet address and hostname

• Hard drive configuration

• Media device names

• Hardware vendor contact information and contract number

• Configuration information for each piece of hardware, both active and inactive, within the organization or organizational site

To obtain hardware information for the following operating systems:

• NetWare, see

"Obtain the Hard Drive Information" on page 102

• UNIX, see

"Prerequisites" on page 23

• Windows, see

"Replacing a Hard Drive" on page 55

18

NetWorker V7.3

Disaster Recovery Guide

Chapter 2: Preparing for a Disaster

Software Information

Maintain the following up-to-date information regarding computer software:

• Copies of the original operating system media and patches (and where they are located)

• Software licenses, enabler and authorization codes

• Software vendor contact information and contract number

• The operating system version and patches installed

• Operating system configuration

• Emergency media that can be used to recover a computer if a disaster occurs

• NetWorker bootstrap information for each NetWorker server

• Kernel configuration and location

• Device drivers

• A list of any Windows volume mount points

Table 1 on page 19

lists where to obtain software information for a specific operating system.

Table 1. Operating System Information (Part 1 of 2)

Platform

UNIX

Windows

Recovery/Repair Prerequisite Information

Recovering the UNIX operating system

"Reinstalling and

Configuring the UNIX

Operating System" on page 24

Recovering a UNIX NetWorker server

"Recovering a NetWorker

Server" on page 28

Recovering a UNIX NetWorker storage node

"Recovering a NetWorker

Storage Node" on page 38

Recovering a UNIX NetWorker client

"Recovering a NetWorker

Client" on page 42

Windows XP Pro and Windows

2003:

Automated System Recovery

(ASR) requirements

"ASR Recovery

Requirements" on page 57

2

NetWorker V7.3

Disaster Recovery Guide

19

2

Chapter 2: Preparing for a Disaster

Platform

NetWare

Macintosh

Table 1. Operating System Information (Part 2 of 2)

Recovery/Repair Prerequisite Information

Windows NT 4.0:

Repairing a Windows NT 4.0 operating system installation

Windows 2000:

Recovering a Windows installation

Windows 2003 and Windows

2000:

NetWorker server recovery requirements

"Repairing a Windows NT

4.0 Operating System

Installation" on page 61

"Recovering a Windows

Installation" on page 64

"NetWorker Server

Recovery Requirements" on page 71

Windows (all versions):

NetWorker client or storage node recovery requirements

"Recovering a NetWorker

Client or Storage Node" on page 87

Windows (all versions):

Dynamic Host Configuration

Protocol (DHCP) and Windows

Internet Naming Service

(WINS) database recovery

"Recovery of DHCP and

WINS Databases" on page 93

Recovering the NetWare operating system

Recovering a NetWare

NetWorker server

Recovering a Mac OS X

NetWorker client

"Recovering the NetWare

Operating System" on page 103

"Recovering a NetWare

NetWorker Server" on page 107

"Recovering a NetWare

NetWorker Server" on page 107

20

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

This chapter explains how to recover from a disaster on a UNIX server, client, or storage node host computer running NetWorker V7.3 software.

This chapter includes the following sections:

"Disaster Recovery Procedures" on page 22

"Replacing a Hard Drive" on page 23

"Reinstalling and Configuring the UNIX Operating System" on page 24

"Recovering a NetWorker Server" on page 28

"Recovering a NetWorker Storage Node" on page 38

"Recovering a NetWorker Client" on page 42

"Recovering the NetWorker Management Console Server" on page 45

NetWorker V7.3

Disaster Recovery Guide

21

3

Chapter 3: UNIX Disaster Recovery

Disaster Recovery Procedures

Table 2 on page 22 lists the disaster recovery procedures that you may have to

perform depending on the damage caused by the disaster. Complete each procedure, as necessary, in the order listed in this table.

Table 2. UNIX Disaster Recovery Procedures

If you are recovering: See:

A hard drive

The operating system

A NetWorker server

A NetWorker storage node

"Replacing a Hard Drive" on page 23

"Reinstalling and Configuring the UNIX

Operating System" on page 24

"Recovering a NetWorker Server" on page 28

"Recovering a NetWorker Storage Node" on page 38

A NetWorker client

"Recovering a NetWorker Client" on page 42

Data* on any computer Refer to the NetWorker Administrator’s Guide

Data* on a computer with EMC

Celestra Power installed

Refer to the EMC

Celestra Administrator’s

Guide

* Refers to data, such as application and user data, that is not part of the core NetWorker software or operating system programs.

22

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

Replacing a Hard Drive

The following sections provide important information to be aware of when replacing a hard drive as part of a disaster recovery:

"Prerequisites" on page 23

"Replacing the Hard Drive" on page 23

"Completing the Recovery Process" on page 24

Prerequisites

To help ensure that you are prepared to replace and reconfigure a hard drive, maintain a current record of the system information. Obtain the following information by using the appropriate operating system commands:

• Size of the drive

• File-system volume information

• Volume label assigned to each disk partition

• How the disk is partitioned

• How the disk is loaded

• The size of the disk

• Each logical volume (size and label)

• Each file system

Note:

Though it does not affect NetWorker operation, you could also note any use of mirroring, Redundant Array of Independent Disk (RAID), striping, compression, or volume sets.

Replacing the Hard Drive

If one or more hard drives fail, refer to the appropriate operating system documentation and hard drive vendor documentation for detailed instructions on how to replace the hard drives.

Important:

To ensure that you can recover all of the drive’s data, install a new drive that is the same size or larger than the original drive.

3

NetWorker V7.3

Disaster Recovery Guide

23

3

Chapter 3: UNIX Disaster Recovery

Completing the Recovery Process

To complete the recovery process after replacing the hard drive, recover the following:

1. The UNIX operating system, if necessary. For information, see

"Reinstalling and Configuring the UNIX Operating System" on page 24 .

2. The NetWorker software corresponding to the type of NetWorker computer you are recovering. For more information, see:

"Recovering a NetWorker Server" on page 28 .

"Recovering a NetWorker Storage Node" on page 38 .

"Recovering a NetWorker Client" on page 42 .

Each of the preceding procedures describe how to recover the NetWorker software as well as the computer’s application and user data.

Reinstalling and Configuring the UNIX Operating System

The following sections provide information on reinstalling and reconfiguring the UNIX operating system:

"Prerequisites" on page 24

"How to Reinstall and Configure the UNIX Operating System" on page 25

"Completing the Recovery Process" on page 28

Prerequisites

To recover the operating system, record the following information and ensure that it is current:

• Version and patch level of the operating system

• Installation path of the operating system

• TCP/IP properties:

– Adapter type

– IP address

– Default gateway

– Subnet mask

– DNS server

24

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

• Computer properties:

– Hostname

– DNS domain name

– Superuser password

• Device and SCSI drivers

• Boot files required for booting the kernel. For example:

– /unix

– /boot

– /etc/default/boot

– /stand/vmunix

DNS Server

If the host being recovered uses DNS for hostname resolution, the DNS server must be available or hostnames cannot be resolved. This could cause the

NetWorker software to be unresponsive when attempting to resolve hostnames.

If the DNS server is not available, disable DNS lookup for the host being recovered. When the DNS server is available, re-enable DNS lookup.

How to Reinstall and Configure the UNIX Operating System

When recovering a UNIX operating system, you must reinstall the same version of UNIX and any patches that were in use prior to the disaster. Then configure only those features that enable the computer to communicate over the network. You can recover the remaining operating system configuration settings after reinstalling the NetWorker software.

You can recover the UNIX operating system back to the original computer or to a different computer.

To recover the operating system, complete the following tasks:

"Task 1: Reinstall the Operating System" on page 26

"Task 2: Configure the Operating System" on page 26

"Task 3: Configure the Devices and Test the Operating System" on page 27

3

NetWorker V7.3

Disaster Recovery Guide

25

Chapter 3: UNIX Disaster Recovery

3

Important:

Do not reinstall the operating system from a UNIX X-Windows session. Instead, recover the operating system in single-user mode from the system console.

26

Task 1: Reinstall the Operating System

To reinstall the operating system:

1. Reinstall the same version and patch level of the UNIX operating system.

Specify the same installation locations that were used in the predisaster installation of the operating system. For installation information, refer to the appropriate UNIX documentation.

To upgrade the operating system, recover the predisaster version of the operating system before performing any upgrades.

2. Re-create all of the file systems that were previously on the computer.

The file systems must be:

• The same file-system type.

• At least the same partition size as the original to hold all of the data that was previously backed up.

At a minimum, configure the root volume group and kernel parameters, such as asynchronous I/O. If the resources are available, reconstruct nonroot volume group file systems and logical volumes.

Task 2: Configure the Operating System

To configure the operating system:

1. Configure the network exactly as it was before the disaster. If you do not, the NetWorker software will treat the computer as a new computer. If the computer has a different host ID, you must reregister the NetWorker software. For information on registering the NetWorker software, refer to the appropriate NetWorker Installation Guide or, if using a Fujitsu Siemens

Computers NetWorker host, the NetWorker Administrator’s Guide.

2. If you are recovering the operating system to a different computer, assign the same hostname to the new computer. Otherwise, you will not be able to recover the NetWorker indexes associated with the original computer.

3. Configure the date and time as they were before, including the time zone.

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

4. Reinstall any additional UNIX components or services before recovering the computer’s data.

5. Reboot the computer.

Task 3: Configure the Devices and Test the Operating System

To configure the devices and test the operating system:

1. Configure any devices the NetWorker software requires, for example SCSI pass-through devices for autochangers. For information about configuring devices, refer to the NetWorker Administrator’s Guide.

2. Test for the following:

• Name to Address resolution and TCP/IP are functioning properly. To test these, run ping on the server to reach a client, and then run ping on a client to reach the server.

• Operating system is functioning properly.

• Tape drive is functioning properly. To test this, run the mt commands.

• Operating system recognizes devices. If the operating system is not recognizing devices, you might need to:

– Load the SCSI driver.

– Reinstall the device driver software.

– Modify the device configuration files to enable the computer to communicate with the device during recovery. For details on modifying the device configuration files, refer to the NetWorker

Administrator’s Guide.

3. Ensure that the block size mode for tape devices used with NetWorker is set to variable. Otherwise, data recovery may fail. The procedure for setting the device block size varies depending on your operating system. For example, to set the tape device block size on an AIX system: a. Open the IBM System Management Interface Tool (SMIT) and select the Devices box in the System Management section near the bottom of the SMIT window.

b. Select Tape Drive.

c. Select Change/Show Characteristics of a Tape Drive. This brings up a list of tape drives.

d. Select a tape drive from the list. The attributes of the selected tape drive will display.

3

NetWorker V7.3

Disaster Recovery Guide

27

3

Chapter 3: UNIX Disaster Recovery e. Ensure that the BLOCK size (0=variable length) (Num.) is set to 0.

f. Repeat step d

and

step e for each tape drive the NetWorker software

uses.

For information about setting the tape device block size on the operating system, refer to the operating system documentation.

Completing the Recovery Process

To complete the recovery process after reinstalling and reconfiguring the operating system, use the following procedures, depending on the type of

NetWorker computer you are recovering. These procedures describe how to recover the NetWorker software as well as the computer’s application and user data:

"Recovering a NetWorker Server" on page 28 .

"Recovering a NetWorker Storage Node" on page 38 .

"Recovering a NetWorker Client" on page 42 .

Recovering a NetWorker Server

The following sections provide information on recovering a NetWorker server:

"Prerequisites" on page 28

"How to Recover a NetWorker Server" on page 29

Prerequisites

Before you can recover the NetWorker server, be sure that the UNIX operating system is reinstalled on the computer. For information about reinstalling the

UNIX operating system, see

"Reinstalling and Configuring the UNIX

Operating System" on page 24 .

To recover the NetWorker server, you need the following:

• Version and patch level of the NetWorker server software.

• Original directory location to which the NetWorker server was installed.

• NetWorker server installation media.

• Backup or clone volumes containing the NetWorker server bootstrap and indexes.

• Name of any links to NetWorker directories. An example of a typical link from a NetWorker directory to a user directory is /nsr to /var/nsr.

28

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

How to Recover a NetWorker Server

This section describes how to recover the NetWorker server back to the original computer.

Important:

If the NetWorker server was also a EMC License Manager server, the EMC License Manager lictype.res file is not recovered when the server is recovered. This is because the EMC License Manager files and directories are not considered part of the NetWorker server. After recovering the NetWorker server, you must explicitly recover the EMC License Manager as a client of the

NetWorker server.

To recover a NetWorker server, complete the following tasks:

"Task 1: Reinstall the NetWorker Server Software" on page 29

"Task 2: Configure NetWorker Device Resources" on page 30

"Task 3: Locate the Server’s Bootstrap Save Set ID" on page 31

"Task 4: Recover the NetWorker Server Bootstrap" on page 32

"Task 5: Rename the NetWorker Server Resource Directory" on page 34

"Task 6: Recover All Client File Indexes" on page 35

"Task 7: Recover the Application and User Data" on page 36

"Task 8: Create a Record of Backups Done after Bootstrap Created" on page 37

"Task 9: Perform a Test Backup and Recovery" on page 38

Task 1: Reinstall the NetWorker Server Software

To reinstall the NetWorker server software:

1. Reinstall the same version of the NetWorker server software into its original location. When you reinstall the NetWorker server software, the

NetWorker client is also installed. For installation instructions, refer to the appropriate NetWorker Installation Guide or, if using a Fujitsu Siemens

Computers NetWorker host, the NetWorker Administrator’s Guide.

Note:

If you want to upgrade the NetWorker server, first recover the server to its original state, then perform the upgrade.

2. Reinstall any NetWorker patches that were installed prior to the disaster.

3

NetWorker V7.3

Disaster Recovery Guide

29

3

Chapter 3: UNIX Disaster Recovery

3 . If the /nsr directory or any of its subdirectories except for /nsr/res were linked, re-create these links. For more information about the /nsr/res directory, see

step 4 .

To re-create links to NetWorker directories: a. Stop the NetWorker daemons by entering the nsr_shutdown command at the command line:

nsr_shutdown -a

b. Move the files to their original location.

c. Create the links from the /nsr directory to the original location.

d. Restart the NetWorker server by first entering the nsrexecd command and then entering the nsrd command, for example:

nsrexecd nsrd

Note:

You can also restart the NetWorker server by running the

NetWorker startup script for the appropriate platform.

4. Determine whether the /nsr/res directory was linked to another directory:

• If no, skip this step.

• If yes, do not re-create the link; instead, create a /nsr/res directory. This directory is used temporarily in

"Task 4: Recover the NetWorker

Server Bootstrap" on page 32 .

Note:

The resource database will not be recovered to the /nsr/res directory created in this step. Instead, the resource directory will be recovered to the link’s target directory. For example, if /nsr/res was linked to a directory named /bigres, then the resource database is recovered to /bigres. Ensure that there is enough space in the target directory for the recovered resource database.

Task 2: Configure NetWorker Device Resources

To configure the NetWorker software:

• Configure the device resources.

– To recover data by using a stand-alone device, ensure that a resource for the stand-alone device exists (this is defined in the /nsr/res directory). If a resource for the stand-alone device does not exist, create it by using the NetWorker Administrator program.

30

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

• To recover data by using an autochanger, ensure that an autochanger resource exists (this is defined in the /nsr/res directory). If the

Autochanger resource does not exist, create it using the NetWorker

Console Administration interface. For details about the NetWorker

Console Administration interface, refer to the NetWorker Administrator’s

Guide.

– Reset the autochanger by using the nsrjb -vHE command. This command resets the autochanger, ejects backup volumes, reinitializes the element status, and checks each slot for a volume.

a. If the autochanger does not support the -E option, initialize the element status by using sjiielm (on Linux, use ielem).

b. Inventory the autochanger by using the nsrjb -I command. This helps to determine whether the volumes required to recover the bootstrap are located inside the autochanger.

Task 3: Locate the Server’s Bootstrap Save Set ID

The bootstrap contains the media database and the resource database.

Important:

If you routinely move NetWorker backup media to an offsite location for safekeeping, and a subsequent file recover operation generates a mount request, the recover operation waits until an operator mounts the request media. To avoid delays when recovering files, use the mminfo -mv command to list the media that is associated with the file you want to recover and to retrieve the media from the offsite storage before starting the recover.

To locate the save set ID of the most recent bootstrap:

1. Insert the most recent media or clone volumes used for scheduled backups into the appropriate device.

2. At the command line, switch to the directory where the NetWorker binaries and executables are located.

3

NetWorker V7.3

Disaster Recovery Guide

31

3

Chapter 3: UNIX Disaster Recovery

3. If you are using an autochanger, insert the first volume of the bootstrap save set into the first drive of the autochanger by using the following command:

nsrjb -lnv -S

slot

-f

device_name

where:

slot is the slot where the first volume is located.

device_name is the pathname for the first drive. You can obtain the

device_name by using the inquire command.

4. Use the scanner -B command to determine the save set ID of the most recent bootstrap on the media. For example:

Solaris

:

scanner -B

/dev/rmt/0hbn

Linux

:

scanner -B

/dev/nst0

If you do not locate the save set ID of the most recent bootstrap on the most recent media, run the scanner -B command on preceding media to locate the save set ID of the most recent bootstrap.

5. Record both the bootstrap save set ID and the volume label from the output.

Task 4: Recover the NetWorker Server Bootstrap

You can use the mmrecov command to recover the NetWorker server bootstrap.

Note:

In NetWorker V6.0 and higher, the mmrecov command is only used to recover the NetWorker server’s media database and resource database; the

nsrck

command is used to recover the server’s client file indexes.

To recover the NetWorker server bootstrap:

1. Use the mmrecov command to recover the NetWorker server’s bootstrap

(media database and resource database). For example:

mmrecov

Output similar to the following appears: mmrecov: Using madrid.spain.com as server

32

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

Important:

The mmrecov command overwrites the server’s media database.

However, mmrecov does not overwrite the resource database. Instead,

mmrecov

recovers the resource database to an alternate location.

If the predisaster resource database was located in the default directory, which is /nsr/res, it will be recovered to /nsr/res.R.

If the predisaster resource database was linked, then the resource database will be recovered to the link’s target directory. For example, if /nsr/res was linked to a directory named /bigres, then the resource database will be recovered to a directory named /bigres.

2. If the server has multiple devices configured and enabled, the following message appears. If you receive this message, enter the name of the device you are using for the recovery.

What is the name of the device you plan on using

[/dev/rmt/0hbn]? /dev/rmt/0hbn

3. When the following message appears, enter the save set ID for the latest bootstrap. If you are recovering a cloned version of the bootstrap, specify the save set ID associated with the clone.

Enter the latest bootstrap save set ID []: 20076

4. When the following message appears, enter the file number to begin the recovery. If unknown, press [Return].

Enter starting file number (if known) [0]: 130

5. When the following message appears, enter the first record number to begin the recovery. If unknown, press [Return].

Enter starting record number (if known) [0]: 0

6. When the following message appears, follow the prompt:

Please insert the volume on which save set id 20076 started into /dev/rmt/0hbn.

Once you have loaded the appropriate volume, the following message appears:

Scanning /dev/rmt/0hbn for save set 20076; this might take a while...

3

NetWorker V7.3

Disaster Recovery Guide

33

3

Chapter 3: UNIX Disaster Recovery

NetWorker then scans the volume for the appropriate save set and recovers it. The NetWorker media database and resource database are recovered when the following message appears:

If your resource files were lost, they are now recovered in the ‘res.R’ directory. Copy or move them to the ‘res’ directory, after you have shut down the service. Then restart the service.

Otherwise, just restart the service.

If the on-line index for the server-name was lost, it can be recovered using the nsrck command.

Task 5: Rename the NetWorker Server Resource Directory

Because the resource files cannot be reliably overwritten while the NetWorker software is running, mmrecov recovered the res directory to an alternate location.

To rename the resource directory:

1. Stop the daemons by entering the nsr_shutdown command at the command line:

nsr_shutdown -a

2. Rename the existing /nsr/res directory to /nsr/res.orig:

mv res res.orig

3. Access the recovered resource database.

If the resource database was recovered to /nsr/res.R, rename /nsr/res.R to

/nsr/res:

mv res.R res

If the /nsr/res directory was linked before the disaster, then the resource database is recovered to the link’s target directory. For example, if /nsr/res was linked to a directory named /bigres, then the resource database is now recovered to /bigres. In this case, re-create the link from the /nsr/res directory to the target directory. For example:

ln -s

/bigres /nsr/res

34

NetWorker V7.3

Disaster Recovery Guide

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

4. Restart the NetWorker server by first entering the nsrexecd command, then entering the nsrd command, for example:

nsrexecd nsrd

Note:

You can also restart the NetWorker server by running the NetWorker startup script for the appropriate platform.

5. After verifying that the NetWorker resources are correct, remove the

/nsr/res.orig directory.

6. Run the nsrjb -HE command to reset the autochanger.

7. If you are using an autochanger, run the nsrjb -Iv command to reinventory the autochanger; or run the nsrjb -Iv -S command to reinventory only the affected slots.

8. If the NetWorker Console server was installed on the same machine as the

NetWorker server, re-install the console software. For installation instructions, refer to the NetWorker Administrator’s Guide. After reinstalling the NetWorker Console server, add the NetWorker server to list of hosts managed by the NetWorker Console server. For instructions on adding the server to the list of hosts, see the NetWorker Administrator’s Guide.

Task 6: Recover All Client File Indexes

Once you recover the server’s media database and resource database, recover all client file indexes on the NetWorker server. You will recover one client file index for each NetWorker client that the NetWorker server backed up.

Note:

Because a NetWorker server always has the NetWorker client installed, this task includes recovering the client file index for the NetWorker server.

To recover all client file indexes:

1. Enter the nsrck -L7 command:

nsrck -L7

client_name

2. If you are using a cloned version of the client file index, the NetWorker server may prompt you to load an original volume (not the clone volume).

To use the clone: a. Press [Ctrl]+[c] to exit out of nsrck and verify that the pending original volume message has terminated. b. Delete the records of the original volumes by using the nsrmm -d

volume_name command. For example:

35

3

3

Chapter 3: UNIX Disaster Recovery

The NetWorker server requests the original volume mars.1. Because this volume is not available, delete the mars.1 volume from the media database by using the following command:

nsrmm -d mars.1

c. Enter the nsrck -L7 command:

nsrck -L7

client_name

Note:

Although you must recover a computer’s client file index before you can select individual files to recover, you can recover an entire save set on a computer by using a save set recover.

Task 7: Recover the Application and User Data

To recover the application and user data that was on the NetWorker server:

1. Log in as root.

2. Load and inventory the devices. This ensures that the NetWorker server can recognize the location of each volume.

Note:

If you load a clone volume, you must either delete the original volume from the media database or mark it as suspect in the media database. If you are using a clone volume, it will be used for the remainder of the recovery process.

3 . Run the nwrecover program.

4. Mark all of the directories or files to be recovered.

36

Important:

Overwriting operating system files may cause unpredictable results. The following list contains examples of operating system files that you should consider not overwriting. To determine all of the operating system files that should not be overwritten during a recovery, refer to the operating system documentation.

• Operating system boot files, such as:

– /unix

– /boot

– /etc/default/boot

– /stand/vmunix

• Other operating system files. For example, some of the files not to overwrite on the Solaris platform include:

– /dev directory

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

– /etc/path_to_inst

– /etc/name_to_major

– /etc/mnttab

– /etc/dfs/sharetab

– /etc/rmtab

– /kernel

– /usr/kernel

– /etc/saf/zsmon/_pmpipe

– /etc/sysevent/piclevent_door

Instead of overwriting these files, use the rename or relocate options during recovery. For more information about renaming files during recovery, refer to the NetWorker Administrator’s Guide.

5. Click Start to begin the recovery.

Note:

By default, the directed recover option is enabled when you install the

NetWorker client. If the directed recover option was disabled on the client before the disaster, you need to explicitly set this option to disable directed recoveries to this client. This option is set through the nsradmin command. For more information about setting the disable directed recover option on a

NetWorker client, refer to the NetWorker Administrator’s Guide.

Task 8: Create a Record of Backups Done after Bootstrap Created

After a disaster recovery, the media database will not contain information about any backups done after the bootstrap was created. The media database retains information about the last known file/record marks on the tape, and if backups were done after the bootstrap was created, the file/record information will not be current. For any tapes that contain backups that were completed after the bootstrap was created, use the scanner command with, at a minimum, the -m option to scan the save sets and/or tapes into the media database. This will ensure that the media database has a record of all completed save sets and that valid backups will not be overwritten by future backups.

3

NetWorker V7.3

Disaster Recovery Guide

37

3

Chapter 3: UNIX Disaster Recovery

Additionally, it is recommended that you write protect all appendable volumes until you can complete the scanner activity to prevent scheduled backups from using these volumes. Use the following commands in order to protect these tapes:

1. To identify only appendable volumes, enter the following at the command line:

$

mminfo -q '!full' -r

"volume,client,mediarec,mediafile,next,name"

2. Mark a volume read-only (without confirmation):

$

nsrmm –y –o readonly

volume name

Repeat Steps 3 and 4 for every appendable volume identified in Step 1.

3. Mount the tape into a tape drive:

nsrjb -ln -f

device

Repeat for every appendable volume identified in step 1.

4. Begin scanning in the missing save sets. Use the last file and record number from the previous mminfo command and issue the following scanner command:

$

scanner –f

number –r number -m device

5. If the NetWorker Management Console server was installed on the host and the Console database was lost, recover the database. For instructions, see

"How to Recover the Console Server Database in a Standalone

Environment" on page 46 .

Task 9: Perform a Test Backup and Recovery

To test the server recovery process:

1. Perform a test backup or recovery to make sure the server is fully recovered.

2. Verify that the server and its associated clients are included in a scheduled backup.

Recovering a NetWorker Storage Node

The storage node allows you to access the volumes on which the backups for all of the network computers reside. If the storage node experiences a disaster, it must be recovered before you can recover application and user data to other computers on the network.

38

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

The following sections provide information on recovering a NetWorker storage node:

"Prerequisites" on page 39

"How to Recover a NetWorker Storage Node" on page 39

Prerequisites

Before recovering the NetWorker storage node, ensure that the UNIX operating system is installed on the computer and that the NetWorker server is functioning and available on the network.

If you need to reinstall the UNIX operating system, see

"Reinstalling and

Configuring the UNIX Operating System" on page 24 .

If you need to recover the NetWorker server, see

"Recovering a NetWorker

Server" on page 28 .

Additionally, ensure that you have the following:

• Hostname of the NetWorker server.

• Version and patch level of the NetWorker storage node software that was on the computer before the disaster occurred.

• Name of any links to NetWorker directories. An example of a typical link from a NetWorker directory to a user directory is /nsr to /var/nsr.

How to Recover a NetWorker Storage Node

This section describes how to recover a NetWorker storage node computer back to the original computer or to a different computer.

To recover a NetWorker storage node, complete the following tasks:

"Task 1: Reinstall the NetWorker Storage Node" on page 39

"Task 2: Recover the Application and User Data" on page 40

"Task 3: Perform a Test Backup and Recovery" on page 42

Task 1: Reinstall the NetWorker Storage Node

To reinstall the NetWorker storage node software:

1. Reinstall the same version of the NetWorker storage node software into its original location.

Note:

If you want to upgrade the storage node software, first recover the storage node to its original state, and then perform the upgrade.

NetWorker V7.3

Disaster Recovery Guide

39

3

3

Chapter 3: UNIX Disaster Recovery

2. Reinstall any NetWorker backup utility patches that were installed prior to the disaster.

3. Re-create any links to NetWorker directories.

4. Optionally, use the nwrecover program to perform a test recovery to ensure that the recovery process is functioning properly.

Note:

The NetWorker client software is also installed when you install the storage node software.

The storage node can now access volumes that contain backups for other computers on the network. These volumes contain the application and user data that are required to fully recover computers that were protected with the

NetWorker client software.

If you want to recover the application and user data for other NetWorker clients, see

"Recovering a NetWorker Client" on page 42 .

If you want to recover application and user data that was on the storage node computer, see,

"Task 2: Recover the Application and User Data" on page 40 .

Task 2: Recover the Application and User Data

To recover the application and user data that was on the NetWorker storage node computer:

1. If you need to determine which volumes contain the application and user data backups for this computer, use the mminfo -avot command on the

NetWorker server, for example:

mminfo

-avot -c

storage_node_name

where storage_node_name is the hostname of the computer whose application and user data you are recovering.

2. Start the nwrecover program on the storage node computer.

3. Mark all of the directories and files to be recovered.

Important:

Overwriting operating system files may cause unpredictable results. The following list contains examples of operating system files that you should consider not overwriting. To determine all of the operating system files that should not be overwritten during a recovery, refer to the operating system documentation.

40

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

• Operating system boot files, such as:

– /unix

– /boot

– /etc/default/boot

– /stand/vmunix

• Other operating system files. For example, some of the files not to overwrite on the Solaris platform include:

– /dev directory

– /etc/path_to_inst

– /etc/name_to_major

– /etc/mnttab

– /etc/dfs/sharetab

– /etc/rmtab

– /kernel

– /usr/kernel

– /etc/saf/zsmon/_pmpipe

– /etc/sysevent/piclevent_door

Instead of overwriting these files, use the rename or relocate options during recovery. For more information about renaming files during recovery, refer to the NetWorker Administrator’s Guide.

4. From the Options menu, select Recover Options to set the recover options.

5. In the Recover Options dialog box, selecting Overwrite Existing File.

Important:

If you did not set the recover options, you must select the

Overwrite Existing File option when the Naming Conflict dialog box appears.

To enable automatic overwriting of files with the same name, select the

Suppress Further Prompting option in the Naming Conflict dialog box.

6. Select Start.

7. Reboot the computer when the recovery is complete. The computer should now be recovered.

3

NetWorker V7.3

Disaster Recovery Guide

41

3

Chapter 3: UNIX Disaster Recovery

Note:

By default, the directed recover option is enabled when you reinstall the

NetWorker client. To disable directed recoveries to this client, set the client’s

disable directed recover

option. This option is set through the nsradmin command. For more information about setting the disable directed recover option on the NetWorker client, refer to the NetWorker Administrator’s Guide.

Task 3: Perform a Test Backup and Recovery

To test the NetWorker storage node backup and recovery process:

1. Perform a test backup by using each of the NetWorker backup utilities incorporated into the backup solution.

2. Perform a test recovery by using the NetWorker utility used to back up the data. For details, refer to the NetWorker Administrator’s Guide.

Recovering a NetWorker Client

The following sections provide information on recovering a NetWorker client:

"Prerequisites" on page 42

"How to Recover a NetWorker Client" on page 43

Prerequisites

Before recovering the NetWorker client, ensure that the UNIX operating system is installed on the computer and that the NetWorker server is functioning and available on the network.

If you need to reinstall the UNIX operating system, see

"Reinstalling and

Configuring the UNIX Operating System" on page 24 .

If you need to recover the NetWorker server, see

"Recovering a NetWorker

Server" on page 28 .

Additionally, ensure that you have the following:

• Version and patch level of the NetWorker client.

• The name of the computer on which the NetWorker server is running.

• Name of any links to NetWorker directories. An example of a typical link from a NetWorker directory to a user directory is /nsr to /var/nsr.

42

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

How to Recover a NetWorker Client

This section describes how to recover a NetWorker client computer back to the original computer or to a different computer.

To recover a NetWorker client, complete the following tasks:

"Task 1: Reinstall NetWorker Client" on page 43

"Task 2: Recover the Application and User Data" on page 43

"Task 3: Perform a Test Backup and Recovery" on page 45

Task 1: Reinstall NetWorker Client

To reinstall the NetWorker client software:

1. Reinstall the same version of the NetWorker client into its original location.

Note:

If you want to upgrade the client software, first recover the client to its original state, and then perform the upgrade.

2. Reinstall any NetWorker backup utility patches that were installed prior to the disaster.

3. Re-create any links to NetWorker directories.

4. (Optional) Use the nwrecover program to perform a test recovery to ensure that the recovery process is functioning properly.

Task 2: Recover the Application and User Data

To recover the application and user data:

1. If you need to determine which volumes contain the application and user data backups for this computer, use the mminfo -avot command on the

NetWorker server, for example:

mminfo

-avot -c

client_name

where client_name is the hostname of the computer whose application and user data is to be recovered.

3

NetWorker V7.3

Disaster Recovery Guide

43

3

Chapter 3: UNIX Disaster Recovery

2. Start the nwrecover program on the client computer whose application and user data you are recovering.

3. Mark all of the directories and files to be recovered.

Important:

Overwriting operating system files may cause unpredictable results. The following list contains examples of operating system files that you should consider not overwriting. To determine all of the operating system files that should not be overwritten during a recovery, refer to the operating system documentation.

• Operating system boot files, such as:

– /unix

– /boot

– /etc/default/boot

– /stand/vmunix

• Other operating system files. For example, some of the files not to overwrite on the Solaris platform include:

– /dev directory

– /etc/path_to_inst

– /etc/name_to_major

– /etc/mnttab

– /etc/dfs/sharetab

– /etc/rmtab

– /kernel

– /usr/kernel

– /etc/saf/zsmon/_pmpipe

– /etc/sysevent/piclevent_door

Instead of overwriting these files, use the rename or relocate options during recovery. For more information about renaming files during recovery, refer to the NetWorker Administrator’s Guide.

4. From the Options menu, select Recover Options to set the recover options.

5. In the Recover Options dialog box, select Overwrite Existing File.

44

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

Important:

If you did not set the recover options, you must select the

Overwrite Existing File option when the Naming Conflict dialog box appears.

To enable automatic overwriting of files with the same name, select the

Suppress Further Prompting option in the Naming Conflict dialog box.

6. Select Start.

7. When the recovery is complete, reboot the computer.

The computer should now be recovered.

Note:

By default, the directed recover option is enabled when you install the

NetWorker client. If the directed recover option was disabled on the client before the disaster, you need to explicitly set this option to disable directed recoveries to this client. This option is set through the nsradmin command. For more information about setting the disable directed recover option on a

NetWorker client, refer to the NetWorker Administrator’s Guide.

Task 3: Perform a Test Backup and Recovery

To test the NetWorker client backup and recovery process:

1. Perform a test backup by using each of the NetWorker backup utilities incorporated into the backup solution.

2. Perform a test recovery by using the NetWorker utility used to back up the data. For details, refer to the NetWorker Administrator’s Guide.

Recovering the NetWorker Management Console Server

To recover from the NetWorker Management Console server after a disaster, perform three steps:

1. Reinstall NetWorker software, or recover the software using the disaster recovery procedures described earlier in this chapter.

2. Reinstall Console server software, or recover the software using normal recovery procedures as described in the NetWorker Administrator’s Guide.

3. Recover the NetWorker Management Console server database.

3

NetWorker V7.3

Disaster Recovery Guide

45

3

Chapter 3: UNIX Disaster Recovery

During recovery of the NetWorker Management Console server database, no

NetWorker Console graphical interface will be available. As a result, any messages such as mount requests will not be able to be addressed from the

Console.

To resolve this issue, use the nsrwatch command to view messages, and use the appropriate commands, such as nsrjb, to address the issues. For more information about nsrwatch, nsrjb, and other NetWorker commands, refer to the Command Reference Guide.

How to Recover the Console Server Database in a Standalone

Environment

If the NetWorker Management Console server database becomes corrupted or if the Console server is damaged, use this procedure to recover lost data. When a disaster recovery is necessary, the NetWorker client software must be reinstalled or recovered before recovering the Console database.

For information about backing up the Console database, please see the

NetWorker Administrator’s Guide.

To recover the Console server database:

1. Stop the Console server if it is currently running.

2. Edit the LD_LIBRARY_PATH environment variable:

LD_LIBRARY_PATH=$LD_LIBRARY_PATH:

Console_Install_Dir

/sybasa/lib export LD_LIBRARY_PATH

Where Console_Install_Dir is the directory where the Console software has been installed.

3. At the command prompt, enter the recoverpsm command:

recoverpsm

recover options

Where recover options can be one of the following:

-s

— Specifies the name of the NetWorker server.

-c

— Specifies the name of the NetWorker client.

-f

— Overwrites the existing database file, if one is present.

-t

— Specifies the point in time of the backup that will be recovered.

If this is not specified, the most recent backup is used.

46

NetWorker V7.3

Disaster Recovery Guide

Chapter 3: UNIX Disaster Recovery

-d

— Specifies the destination directory where the database files will be recovered. If this is not specified, the database is restored to the current Console server database directory.

-S

— If the Console server has been moved to a different machine after the last backup, use this option to specify the name of the previous Console server host. The format of the database server name is

gst_on_

<NMC server host name>. <NMC server host name> should be the machines "short name;" for example, if the machine name is

wolf.legato.com, the proper format for this option would be

-S gst_on_wolf

.

4. Restart the Console server.

For more information about the recoverpsm command, see the recoverpsm man page.

How to Recover the Console Server Database in a Cluster Environment

If the Console server was configured as a highly available service, use the following steps to recover the Console server database:

1. On the NetWorker server where the database was backed up.

Ensure that you have properly configured the Console virtual client for remote access where the recover operation will be run: a. Start the NetWorker software.

b. From the Administration window, click Configuration.

c. In the left pane, select Clients. d. From the File menu, select Properties and then select the Globals 2 of

2 tab.

e. Make sure that the Remote Access attribute of the Console virtual client for which you want to recover the data contains the name of the physical host you are using to recover the data. For example:

[email protected]_hostname

2. Stop the Console virtual service using the cluster software.

3. On the cluster node where recovery will be run: a. Mount the shared disk directory where the Console database is to be recovered.

b. Edit the LD_LIBRARY_PATH environment variable:

LD_LIBRARY_PATH=$LD_LIBRARY_PATH:

Console_Install_Dir

/sybasa/lib

NetWorker V7.3

Disaster Recovery Guide

47

3

3

Chapter 3: UNIX Disaster Recovery

export LD_LIBRARY_PATH

Where Console_Install_Dir is the directory where the Console software has been installed.

c. Bring the virtual host online.

d. Run the following command to recover the database:

Console_Install_Dir

/bin/recoverpsm -s

NetWorker_Servername

-d

shared_disk_mnt_pnt

/lgto_gstdb -c

Console_Virtual_Servicename

where:

Console_Install_Dir is the directory where the Console software has been installed.

NetWorker_Servername is the name of the NetWorker server.

shared_disk_mnt_pnt is the shared mount point.

Console_Virtual_Servicename is the servicename of the Console virtual service.

e. Unmount the shared disk directory.

f. Restart the Console service using the cluster software.

For more information about the recoverpsm command, see the recoverpsm man page.

48

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

This chapter explains how to recover from a disaster on a host computer in the following configurations:

• NetWorker V7.x server or storage node on Microsoft Windows 2003 or

Windows 2000

• NetWorker V7.x client on Microsoft Windows 2003, Windows XP

Professional, Windows 2000, or Windows NT 4.0

Note:

The term autochanger refers to a variety of robotic data storage devices, including autoloaders, carousels, datawheels, jukeboxes, and near-line storage.

Although a Microsoft Removable Storage library can be either an autochanger or a stand-alone drive, the NetWorker software provides optional Removable

Storage support for autochangers only. For more information, refer to the

NetWorker Administrator’s Guide.

This chapter includes the following sections:

"Disaster Recovery Summary of Procedures" on page 50

"Disaster Recovery of Multiple Computers" on page 54

"Replacing a Hard Drive" on page 55

"Performing a Windows ASR Recovery" on page 55

"Repairing a Windows NT 4.0 Operating System Installation" on page 61

"Recovering a Windows Installation" on page 64

"Recovering a NetWorker Server" on page 71

"Recovering a NetWorker Client or Storage Node" on page 87

"Recovery of DHCP and WINS Databases" on page 93

"Recovering the NetWorker Management Console Server" on page 94

NetWorker V7.3

Disaster Recovery Guide

49

4

Chapter 4: Windows Disaster Recovery

Disaster Recovery Summary of Procedures

Use the following summary of procedures as a guide to the more detailed procedures required for performing NetWorker disaster recovery on any supported Microsoft Windows platform. Experienced NetWorker administrators may find that these summary procedures are sufficient for most tasks. Users who want more information should refer to the detailed instructions in subsequent sections of this chapter.

Important:

For NetWorker clients running a 32-bit version of Windows, approximately 500 MB of extra disk space is sufficient in most cases. For

NetWorker clients running a 64-bit version of Windows, approximately 1 GB of extra disk space is usually sufficient.

The default location for temporary files is NetWorker_install_path\tmp. For the

VSS SYSTEM save sets, the temporary files are placed in the temp directory on the drive where the original files reside. For information on how to expand the available space in the temp directory, refer to the Microsoft documentation.

When recovering SYSTEM or VSS SYSTEM save sets by using the NetWorker

User program, verify that all save sets were recovered successfully by reviewing the messages in the Recover Status window (or the networkr.log file) after the recovery is complete but before rebooting the client computer.

Summary of a NetWorker Server Recovery

To perform a disaster recovery on a NetWorker server:

1. Replace the hard drive, if necessary. For detailed instructions, see

"Replacing a Hard Drive" on page 55 .

2. Reinstall the operating system, if necessary. For detailed instructions, see

"Recovering a Windows Installation" on page 64 .

Note:

For a guide to the detailed instructions for each of the remaining

steps, see the task list under "Recovering a NetWorker Server" on page 71 .

3. Reinstall the NetWorker software, if necessary.

50

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

4. Configure the storage device. Depending on the configuration, do one of the following:

• If you are using a stand-alone storage device, configure the device and load the volume that contains the latest bootstrap.

• If you are using an autochanger with Removable Storage disabled, perform the following: a. Create a NetWorker Autochanger resource using the NetWorker

Console Administration interface. For details about the

NetWorker Console Administration interface, refer to the

NetWorker Administrator’s Guide.

b. Run nsrjb -vHE to reset the autochanger.

c. Run nsrjb -Iv -S # to inventory the backup tape in slot # (the tape that contains the latest bootstrap).

d. Run nsrjb -lnv -S # -f device_name to load the backup tape in slot # into the device.

• If you are using an autochanger with Removable Storage enabled: a. Use the Microsoft Management Console (MMC) Removable

Storage snap-in to inventory the autochanger.

b. Use the NetWorker Administrator program to create a Device resource for each drive in the autochanger.

c. Use the Removable Storage MMC snap-in to mount the tape that contains the bootstrap.

Note:

If the storage device required for a NetWorker server recovery is connected to a remote storage node, you might have to partially recover the storage node before you can recover the NetWorker server. For more information, see

"Partial Recovery of a NetWorker Storage Node" on page 88 .

5. If you know the bootstrap save set ID, recover the NetWorker media database and resource configuration files by running the mmrecov command from the command prompt.

If you do not know the bootstrap save set ID, do the following: a. Look in the savegrp.log file, or run the scanner -B device_name command from the command prompt to determine the bootstrap save set ID.

b. Run the mmrecov command to recover the NetWorker media database and resource configuration files.

6. Stop the NetWorker services.

NetWorker V7.3

Disaster Recovery Guide

51

4

4

Chapter 4: Windows Disaster Recovery

7. Copy the contents of the <NetWorker_install_path>\res.R directory to the

<NetWorker_install_path>\res directory.

8. Restart the NetWorker services.

9. If you are using an autochanger, reset and inventory it. If you are using a stand-alone drive, skip this step and proceed to

step 10 .

• If Removable Storage is disabled, do the following: a. Run the nsrjb -vHE command to reset the autochanger.

b. Run the nsrjb -Iv command to inventory the autochanger.

• If Removable Storage is enabled, do the following: a. In the NetWorker Console Administration interface, delete the

Removable Storage jukebox through the Autochangers window.

a. Create a NetWorker Autochanger resource using the NetWorker

Console Administration interface. For details about the

NetWorker Console Administration interface, refer to the

NetWorker Administrator’s Guide.

b. In the NetWorker Console Administration interface, use the

Autochanger Operations dialog box to reallocate all of the

NetWorker volumes from the Import pool.

c. In the NetWorker Console Administration interface, click Reset in the Autochanger Operations dialog box. d. Once the reset is complete, click Inventory.

10. Recover the client file indexes by running the nsrck -L7 command from the command prompt.

11. Recover the client’s data and SYSTEM or VSS SYSTEM save sets by using the NetWorker User program.

12. Reboot the computer after the recovery is complete.

52

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

13. If you are using an autochanger with Removable Storage enabled: a. Use the Removable Storage MMC snap-in to mount the tape that contains the bootstrap.

b. Run mmrecov to recover the NetWorker media database and resource configuration files.

c. Stop the NetWorker services.

d. Rename <NetWorker_install_path>\res.R to

<NetWorker_install_path>\res.

e. Start the NetWorker services.

f. On the Autochanger Operations dialog box, click Reset to reset the jukebox.

14. Verify the recovery.

Summary of a NetWorker Client or Storage Node Recovery

In most cases, the NetWorker server should be fully functional before you begin recovery of its associated clients or storage nodes. However, if the storage device required for a NetWorker server recovery is connected to a remote storage node, you might have to partially recover the storage node before you can recover the NetWorker server. For more information, see

"Partial Recovery of a NetWorker Storage Node" on page 88 .

To perform a disaster recovery on a NetWorker client or storage node:

1. Replace the hard drive, if necessary. For instructions, see

"Replacing a

Hard Drive" on page 55 .

2. To begin the software recovery, do one of the following, depending on the

Windows version:

• For Windows 2003 and Windows XP Professional NetWorker clients only, perform an ASR recovery if possible. This requires a current

ASR disk or ASR backup from which you can create an ASR disk. For instructions, see

"Performing a Windows ASR Recovery" on page 55

.

If you use the ASR recovery method, you do not need to perform any of the remaining steps in this summary procedure.

• For Windows 2000 NetWorker clients (or Windows 2003 or Windows

XP Professional clients for which an ASR disk or save set is not available), reinstall the operating system if necessary. For instructions, see

"Recovering a Windows Installation" on page 64 .

4

NetWorker V7.3

Disaster Recovery Guide

53

4

Chapter 4: Windows Disaster Recovery

• For Windows NT 4.0 NetWorker clients, repair the operating system if necessary. For instructions, see

"Repairing a Windows NT 4.0

Operating System Installation" on page 61 .

3. Ensure that the recovery requirements are satisfied. For instructions, see

"Task 1: Satisfy the NetWorker Client or Storage Node Recovery

Requirements" on page 88 .

4. Reinstall the NetWorker software, if necessary. For instructions, see "Task

2: Reinstall the NetWorker Client or Storage Node Software" on page 89 .

5. Recover the data and SYSTEM or VSS SYSTEM save sets by using the

NetWorker User program. For instructions, see

"Task 3: Recover the

NetWorker Client or Storage Node Data" on page 89

.

6. Reboot the computer after the recovery is complete.

7. Verify the recovery. For instructions, see "Task 4: Verify the NetWorker

Client or Storage Node Recovery" on page 92 .

Disaster Recovery of Multiple Computers

If you are performing disaster recovery on more than one computer, recover the computers in the following order:

1. The NetWorker server

2. The NetWorker storage nodes (if any) used by the NetWorker server

3. The NetWorker clients associated with the NetWorker server

Note:

If the backup device to be used for the recovery is connected to a storage node, you might have to perform a partial recovery of the storage node before you begin recovery of the NetWorker server. For more information, see

"Recovering a NetWorker Client or Storage Node" on page 87

.

Disaster Recovery of Clustered Computers

For disaster recovery of a computer that was running NetWorker software in a cluster environment, see the appropriate chapter for procedures:

"Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)" on page 123

"Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and

2003)" on page 139

54

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Replacing a Hard Drive

Important:

To ensure that you have enough disk space to recover all of the failed hard drive’s data, install a new hard drive with at least as much storage capacity as the original drive.

Recovery of the SYSTEM or VSS SYSTEM save sets requires extra disk space for temporary files. The recovery operation might require as much extra disk space as the total size of the SYSTEM or VSS SYSTEM save sets. For an estimate, run mminfo from the command prompt and check the size of the save sets. In many cases, about 500 MB of temporary disk space is sufficient.

To replace the failed hard drive, the following information is required:

• The storage capacity of the hard drive being replaced, plus the amount of extra disk space required for temporary files.

• The drive letter, size, format, and volume label assigned to each partition on the hard drive being replaced:

Windows 2003, Windows 2000, and Windows XP Professional:

This information is available in Control Panel > Administrative Tools >

Computer Management > Disk Management.

Windows NT 4.0:

This information is available in Start > Programs > Administrative Tools

(Common) > Disk Administrator.

Note:

Although it will not affect NetWorker operation, you may also want to note any use of mirroring, RAID/striping, compression, or volume sets.

Install, partition, and format the new hard drive. For instructions, refer to the hard drive manufacturer’s documentation and the appropriate Microsoft documentation.

Performing a Windows ASR Recovery

Microsoft ASR is a feature of the Windows 2003 and Windows XP Professional operating systems. For more information about NetWorker software support for ASR, including procedures for performing ASR backups and creating an

ASR disk, refer to the NetWorker Administrator’s Guide.

NetWorker V7.3

Disaster Recovery Guide

55

4

4

Chapter 4: Windows Disaster Recovery

The ASR recovery method is supported only for NetWorker clients on

Windows 2003 and XP Professional platforms.

ASR recovery is not supported for NetWorker clients operating in a Windows

2003 Microsoft Cluster Server (MSCS) environment.

To recover a NetWorker server or storage node, or NetWorker clients operating in a Windows 2003 MSCS environment, you must use the legacy NetWorker disaster recovery method. For details, see the appropriate procedure:

"Summary of a NetWorker Server Recovery" on page 50

"Summary of a NetWorker Client or Storage Node Recovery" on page 53 .

"Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and

2003)" on page 139

To use the ASR recovery method, perform these tasks in the specified order:

"Task 1: Perform an ASR Recovery of the NetWorker Client Computer" on page 57

"Task 2: Recover Components that Require Special Handling" on page 59

"Task 3: Verify the NetWorker Client Recovery" on page 60

Limitations of the Operating System’s Recovery CD-ROM

Many computer manufacturers, such as Dell, Hewlett-Packard, Compaq, and

IBM, provide a recovery CD-ROM or DVD with each system, which typically includes the Windows operating system installation files and any additional software that was included with the system. These recovery disks cannot be used to perform an ASR recovery.

Important:

To perform an ASR recovery, you must have an official Microsoft

Windows installation CD-ROM for the version of Windows you are recovering. If you have only the recovery CD-ROM supplied by the original equipment manufacturer (OEM) that includes the Windows installation files, skip this section and proceed to

"Recovering a Windows Installation" on page 64 . Refer to the OEM documentation as necessary.

56

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

ASR Recovery Requirements

To perform an ASR recovery of a NetWorker client host computer, you need:

• The ASR disk for the computer you are recovering. If a current ASR disk or ASR save set is not available for the failed NetWorker client host computer, you must use the legacy NetWorker disaster recovery method.

For a guide to the required procedures, see

"Recovering a NetWorker

Client or Storage Node" on page 87 .

• The official Microsoft Windows 2003 or Windows XP Professional installation CD-ROM for the computer you are recovering.

• The latest NetWorker backup for the computer you are recovering.

Note:

To avoid delays during a recovery, if you routinely move NetWorker backup media to an offsite location for safekeeping, ensure that all necessary volumes are available. To list the media associated with the files you want to recover, run mminfo -mv from the command prompt. For more information about the mminfo command, refer to the NetWorker Command Reference Guide.

Task 1: Perform an ASR Recovery of the NetWorker Client Computer

To perform an ASR recovery of the NetWorker client computer:

1. On the NetWorker client computer you are recovering, boot from the

Windows 2003 or Windows XP Professional installation CD-ROM.

Note:

You may need to run the BIOS setup program to configure the computer to boot from the CD-ROM drive. For the procedure, refer to the computer manufacturer’s documentation.

2. Watch closely at the beginning of the boot process. If prompted, press a key to boot from the CD-ROM drive.

3. During the text-mode phase of Windows setup, watch the lower portion of the screen. When prompted, press [F2] to display the ASR Recovery menu. Follow the instructions on the screen.

4. When prompted, insert the ASR disk into drive A: and press a key to continue. ASR formats the system partition, copies files, and begins the

Windows installation.

Note:

Due to a problem with Microsoft Windows XP Professional, when you are prompted to insert the ASR disk and press a key to continue, you may need to press a key several times before the system recognizes the disk and proceeds with the recovery. This problem does not occur on

Windows 2003 systems.

NetWorker V7.3

Disaster Recovery Guide

57

4

4

Chapter 4: Windows Disaster Recovery

5. If you did not select the Pause During Recovery option while creating the

ASR disk, a fully automated recovery will be performed; there will be no pause and you will not be prompted to select which save sets to recover.

If you selected the Pause During Recovery option while creating the ASR disk, during the graphical phase of Windows setup, the NetWorker ASR

Client dialog box appears. Expand My Computer and mark the save sets you want to recover. For example, mark the following save sets:

C:\

D:\

SYSTEM STATE:\

SYSTEM DB:\

SYSTEM FILES:\

If Volume Shadow Copy Service (VSS) is licensed and enabled, mark the following VSS save sets:

VSS SYSTEM BOOT:\

VSS SYSTEM FILESET:\

Note:

VSS USER DATA

,

VSS OTHER

, and

VSS SYSTEM SERVICES

do not appear because they are not required to boot from ASR mode.

By default, the displayed save sets represent the most recent backup. You can view and select previous backups by entering a new browse time in the Browse Time field. The browse time must be entered in time and date the nsr_getdate format; for example, a date can be specified using the format mm/dd/yy or monthname dd, yy. For more information about nsr_getdate, refer to the Command Reference Guide.

6. Select Continue to complete the recovery. ASR finishes installing

Windows, and then automatically runs a NetWorker recovery of the save sets selected in the previous step.

Important:

The VSS components and certain legacy components cannot be correctly recovered during an ASR recovery. To ensure that all necessary components are properly recovered, see

"Task 2: Recover Components that

Require Special Handling" on page 59 .

58

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Task 2: Recover Components that Require Special Handling

Important:

Due to limitations in Microsoft ASR functionality, the following system state components cannot be correctly recovered during an ASR recovery:

– COM+ Registration Database

– Disk Quota Database

– Windows Management Instrumentation Database

– VSS writers

If the NetWorker client being recovered uses any of these components, the following procedures must be performed after an ASR recovery.

If the NetWorker client being recovered does not use any of these components,

skip this task and proceed to "Task 3: Verify the NetWorker Client Recovery" on page 60 .

For NetWorker clients with components that require special handling, do the following after an ASR recovery:

1. Log in with administrator privileges to the target computer.

2. Start the NetWorker User program.

3. Click the Recover toolbar button.

4. In the Source Client dialog box, click OK to select the local client.

5. In the Destination Client dialog box, click OK to select the local client.

6. In the Recover window, select the SYSTEM STATE save set and check for the presence of the COM+ Registration Database component. If it is present, mark the SYSTEM STATE save set for recovery.

Alternatively, if VSS is licensed and enabled, mark all VSS save sets for recovery, except VSS ASR DISK.

4

NetWorker V7.3

Disaster Recovery Guide

59

4

Chapter 4: Windows Disaster Recovery

7. In the Recover window, select the SYSTEM DB save set and check for the presence of the Disk Quota Database and the Windows Management

Instrumentation Database. If they are present, mark the SYSTEM DB save set for recovery.

Note:

If VSS is licensed and enabled, skip this step.

8. If you marked any save sets for recovery, click the Start toolbar button to begin the recovery.

Task 3: Verify the NetWorker Client Recovery

To verify the NetWorker client recovery:

1. Reboot the NetWorker client host computer and verify that the NetWorker

Remote Exec and NetWorker Power Monitor services have started.

Note:

You can disable the NetWorker Power Monitor service if it is not needed. For more information, refer to the NetWorker Administrator’s

Guide.

2. Use the Windows Event Viewer to examine the event logs for errors. In particular, check for the following:

• Service startup errors related to the Windows system state

• Errors regarding the recovery of Windows system-protected files

For information about how the NetWorker software handles the Windows system state and system-protected files, refer to the NetWorker

Administrator’s Guide.

Note:

VSS is unavailable during an ASR recovery. Once an ASR recovery is complete and the system is rebooted, VSS is available for proper recovery of the writers.

3. Verify that any applications (such as Microsoft Office) that were running prior to the disaster have been properly recovered. To verify this, start each application and open a previously saved document.

60

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Repairing a Windows NT 4.0 Operating System Installation

This section explains how to use NetWorker software to create a Windows NT

4.0 Emergency Repair Disk (ERD), and how to use the ERD to repair a damaged Windows NT 4.0 operating system installation.

If you do not have a current ERD, see "Creating an ERD from the REPAIRDISK

Save Set" on page 61 .

If you have a current ERD, see

"How to Use the ERD to Repair the Windows

NT 4.0 Installation" on page 63 .

Windows NT 4.0 Recovery Requirements

To repair a Windows NT 4.0 operating system installation requires:

• Windows NT 4.0 Setup Disks (three disks)

• Windows NT 4.0 installation CD-ROM (same version that was installed prior to the disaster)

• Windows NT 4.0 Service Pack (same level that was installed prior to the disaster)

• For the computer that is being recovered, one of the following:

– A current Windows NT 4.0 ERD.

– The NetWorker backup or clone volume that contains the most recent

REPAIRDISK save set.

• The Administrator account password for the computer to be recovered

(required only if you are repairing the Registry or system state)

Creating an ERD from the REPAIRDISK Save Set

If you do not have a current ERD for the computer you are recovering, you can create one by using either the NetWorker User program or the

<NetWorker_install_path>\bin\recover.exe program.

The operating system of the computer you are recovering needs repair.

Because of this, to create an ERD, you must use a different NetWorker client computer to perform a directed recovery of the REPAIRDISK save set. Before beginning the directed recovery, ensure that:

• The computer used to recover the REPAIRDISK save set is a functional

NetWorker for Windows client, and is connected to the NetWorker server used for the recovery.

4

NetWorker V7.3

Disaster Recovery Guide

61

4

Chapter 4: Windows Disaster Recovery

• The Client resource for the computer to be recovered includes in its

Remote Access list the NetWorker server used for the recovery.

• The NetWorker server used for the recovery includes in its Administrator list the account used to recover the REPAIRDISK save set.

For details about meeting these requirements, refer to the NetWorker

Administrator’s Guide.

How to Use the NetWorker User Program to Create an ERD

To create an ERD by using the NetWorker User program:

1. Start the NetWorker User program.

2. Insert a blank, formatted disk in drive A:.

3. Click the Recover toolbar button.

4. In the Source Client dialog box, select the computer you are recovering and click OK.

5. In the Destination Client dialog box, select the computer you are using to create the ERD and click OK.

6. Mark the REPAIRDISK save set for recovery.

7. From the Options menu, select Recover Options.

8. In the Relocate Recovered Data To dialog box, enter A:\ and click OK.

9. Click the toolbar Start button to begin recovery of the REPAIRDISK save set to the disk in drive A:.

Important:

If the REPAIRDISK save set is larger than 1.44 MB, it will not fit on a disk. This is a known limitation of the Windows NT 4.0 operating system.

How to Use the NetWorker recover Command to Create an ERD

The NetWorker recover.exe program is located in

<NetWorker_install_path>\bin. There is also a Windows system command named recover.exe, located in%SystemRoot%\System32.

62

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Important:

To avoid executing the Windows version of recover.exe when attempting to execute the NetWorker command, do one of the following:

• Before executing the recover.exe command, change to

<NetWorker_install_path>\bin.

• At the command prompt, include the full path and filename to

<NetWorker_install_path>\bin\recover.exe.

• Ensure that <NetWorker_install_path>\bin occurs before%SystemRoot%\System32 in your PATH environment variable.

To create an ERD by using the NetWorker recover command:

1. On the NetWorker server, insert a blank, formatted disk into drive A:.

2. Change to the directory that contains the NetWorker binary files. The default location is <NetWorker_install_path>\bin.

3. Recover the REPAIRDISK save set by using the recover command as follows:

recover -c

failed_hostname

recover> cd / recover> add REPAIRDISK recover> relocate a:\ recover> recover

This procedure initiates recovery of the failed computer’s REPAIRDISK save set to the disk in the NetWorker server’s A: drive.

How to Use the ERD to Repair the Windows NT 4.0 Installation

To use the ERD to repair the Windows NT 4.0 installation:

1. On the computer to be recovered, insert Windows NT 4.0 Setup Disk 1 into drive A: and reboot the computer.

2. When prompted, insert Windows NT 4.0 Setup Disk 2 into drive A: and press [Enter].

3. When the Setup Welcome screen appears, press [R] to select the Repair option.

4

NetWorker V7.3

Disaster Recovery Guide

63

4

Chapter 4: Windows Disaster Recovery

4. Select the appropriate emergency repair options, and then select Continue and press [Enter].

5. When prompted, insert Windows NT 4.0 Setup Disk 3 into drive A: and press [Enter].

6. Insert the ERD into drive A: and press [Enter].

7. Follow the prompts on the screen to repair the Windows files:

• Press [Enter] to repair the specified file.

• Press [Esc] if you do not want to repair the specified file.

8. Reboot the computer.

9. The operating system should now be functional. If the operating system is not functional, see

"Recovering a Windows Installation" on page 64 .

If the operating system appears to be functional:

• Verify that the network protocols are working properly by running the ping command on the NetWorker server’s hostname.

• Verify that the name-to-address resolution is working properly by running the nslookup command on the NetWorker server’s fully qualified domain name.

For more information about the ping and nslookup commands, refer to the Microsoft documentation.

10. If you are using a device that is not directly supported by Windows NT 4.0, you might need to reinstall the device driver software. For information on device driver software, refer to the device manufacturer’s documentation.

Recovering a Windows Installation

This section explains how to recover all supported versions of the Windows operating system during disaster recovery of a NetWorker server, storage node, or client host computer. You can recover the Windows operating system to the original computer or to a different computer.

Note:

For Windows 2003 or Windows XP Professional systems, use the following procedure only as a last resort for recovering a Windows installation. Instead, use the ASR recovery method if possible. For more information, see

"Performing a Windows ASR Recovery" on page 55 .

64

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Windows Recovery Requirements

To recover the Windows operating system, you need to know the following configuration details about the computer immediately prior to the disaster:

• Operating system version and any installed patches, service packs, or option packs

• Path location of the operating system installation

• Host properties:

– Computer name

– Fully-qualified domain name

– Administrator account password

• TCP/IP properties:

– Adapter type

– IP address

– Default gateway

– Subnet mask

– DNS server

– WINS server (if used)

• Any other protocols installed

• Removable Storage enabled/disabled settings

• Virtual memory settings

• Date and time properties

• Localization properties

DNS Server

If the host being recovered uses DNS for hostname resolution, the DNS server must be available or hostnames cannot be resolved. This could cause the

NetWorker software to be unresponsive when attempting to resolve hostnames.

If the DNS server is not available, disable DNS lookup for the host being recovered. When the DNS server is available, re-enable DNS lookup.

4

NetWorker V7.3

Disaster Recovery Guide

65

4

Chapter 4: Windows Disaster Recovery

How to Recover the Windows Operating System

Follow these tasks to recover the operating system, then reinstall and configure only the features that enable the computer to communicate over the network. The other Windows configuration settings are recovered during the

NetWorker recovery.

To recover a Windows installation, perform these tasks in the specified order:

"Task 1: Reinstall the Windows Operating System" on page 66

"Task 2: Configure the Devices and Test the Operating System" on page 69

66

Important:

If you are recovering the operating system to a different computer:

• Assign the same computer name to the new computer that is replacing the original computer; otherwise, you cannot recover data that was backed up from the original computer.

• Assign the same IP address to the new computer that was used for the original computer; otherwise the NetWorker server will assign the computer a new host ID. If the new computer does not have the same IP address as the original computer, you must reregister the NetWorker software. For information on registering NetWorker software, refer to the

EMC NetWorker Installation Guide, Microsoft Windows Version.

Task 1: Reinstall the Windows Operating System

Important:

If you want to upgrade Windows during the recovery, first recover the Windows version that was installed immediately prior to the disaster, then upgrade after you have completed the recovery.

To reinstall the Windows operating system:

1. Reinstall the same version of Windows, including any patches, service packs, or option packs in use immediately prior to the disaster.

During the installation, be sure to set the following configuration options as they were prior to the disaster:

• Windows installation location

• Computer name

• Administrator account password

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

• Regional settings

• Date and time settings

Note:

If the computer that is being recovered was a domain controller prior to the disaster, that configuration will be recovered later in the process.

2. After the operating system is installed, configure the computer as follows:

Windows 2003 and Windows XP Professional:

a. Start the Control Panel System tool.

b. On the System Properties dialog box, click the Computer Name tab and click the Change button.

c. In the Computer Name Changes dialog box, ensure that the computer name is correct and that the computer is a member of the correct domain. Click More.

d. In the DNS Suffix and NetBIOS Computer Name dialog box, ensure that the Primary DNS Suffix Of This Computer text box contains the correct domain name (the name that is appended to the hostname to form the fully-qualified domain name). For example, the fully qualified domain name for a computer named kingdome, might be

kingdome.seattle.washington.com. e. Start the Control Panel Local Area Connection tool.

f. In the Local Area Connection Status dialog box, click Properties.

g. In the Local Area Connection Properties dialog box, select Internet

Protocol (TCP/IP) and click Properties.

h. In the Internet Protocol (TCP/IP) Properties dialog box, set the

TCP/IP properties as they were prior to the disaster, with the same IP address, subnet mask, default gateway, and DNS server.

Windows 2000

: a. Start the Control Panel System tool.

b. In the System Properties dialog box, click the Network Identification tab and click Properties.

c. In the Identification Changes dialog box, ensure that the computer name is correct and that the computer is a member of the correct domain. Click More.

d. In the DNS Suffix and NetBIOS Computer Name dialog box, ensure that the Primary DNS Suffix Of This Computer box contains the correct domain name (the name that is appended to the host name to

4

NetWorker V7.3

Disaster Recovery Guide

67

4

Chapter 4: Windows Disaster Recovery form the fully-qualified domain name). For example, the fully qualified domain name for a computer named kingdome, might be

kingdome.seattle.washington.com.

e. Start the Control Panel NetWork and Dial-up Connections tool.

f. In the Network and Dial-up Connections window, right-click Local

Area Connection and select Properties.

g. In the Local Area Connection Properties dialog box, select Internet

Protocol (TCP/IP) and click Properties.

h. In the Internet Protocol (TCP/IP) Properties dialog box, set the

TCP/IP properties as they were prior to the disaster, with the same IP address, subnet mask, default gateway, and DNS server.

Windows NT 4.0

: a. Start the Control Panel Network tool.

b. In the Network dialog box, click the Identification tab and ensure that the computer name and domain are correct.

c. Click the Protocols tab, select TCP/IP and click Properties.

d. In the Microsoft TCP/IP Properties dialog box, set the TCP/IP properties as they were prior to the disaster, with the same IP address, subnet mask, default gateway, and DNS server.

3 . In the hosts file (%SystemRoot%\system32\drivers\etc\hosts) make an entry for the NetWorker server to be used in the recovery. NetWorker software requires this entry when:

• DNS is not in use or no DNS server is available.

• The NetWorker server host being recovered is also a DNS server.

The entry in the hosts file must include the IP address, the fully-qualified domain name, and the computer name, all on the same line. For example:

123.56.890.474 kingdome.seattle.washington.com kingdome

Note:

If the machine being recovered is also a NetWorker server and no

DNS server is available, each client must also be listed in this file.

4. Configure the virtual memory as it was prior to the disaster:

68

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Windows 2003 and Windows XP Professional:

a. Start the Control Panel System tool.

b. In the System Properties dialog box, click the Advanced tab and click

Settings in the Performance section.

c. In the Performance Options dialog box, click the Advanced tab and click Change in the Virtual Memory section.

d. In the Virtual Memory dialog box, set the virtual memory options.

Click OK.

Windows 2000:

a. Start the Control Panel System tool.

b. In the System Properties dialog box, click the Advanced tab and click

Performance Options.

c. In the Performance Options dialog box, click Change.

d. In the Virtual Memory dialog box, set the virtual memory options.

Click OK.

Windows NT 4.0

: a. Start the Control Panel System tool.

b. In the System Properties dialog box, click the Performance Tab.

c. On the Performance tab, click Change.

d. In the Virtual Memory dialog box, set the virtual memory options.

Click OK.

5. Reinstall any additional Windows components previously installed through Windows Setup, such as Gateway Services for NetWare.

Note:

Do not upgrade OEM drivers for network interface cards (NIC) with the

OEM Service Pack version of the NIC drivers.

Task 2: Configure the Devices and Test the Operating System

If the computer you are recovering is a NetWorker server or storage node, configure any devices required by the NetWorker software.

4

NetWorker V7.3

Disaster Recovery Guide

69

4

Chapter 4: Windows Disaster Recovery

To configure the devices and test the operating system:

1. Configure any devices required by the NetWorker software, for example,

SCSI pass-through devices for autochangers.

2. Ensure that Removable Storage is configured as it was prior to the disaster.

To enable or disable Removable Storage: a. On the desktop, right-click My Computer and select Manage.

b. In the left pane of the Computer Management window, expand

Storage\Removable Storage\Physical Locations.

c. Right-click the icon for the Removable Storage jukebox and select

Properties.

d. If you plan to configure the NetWorker software to use Removable

Storage, select the Enable Library check box in the Properties dialog box.

If you do not plan to configure the NetWorker software to use

Removable Storage, clear the Enable Library check box in the

Properties dialog box.

e. Repeat step c and step d for each stand-alone drive or autochanger

connected to the NetWorker server or storage node host computer.

The NetWorker software supports optional use of Removable Storage for autochangers only. Ensure that Removable Storage is disabled for all stand-alone drives. For more information about Removable Storage, refer to the NetWorker Administrator’s Guide.

3. Verify the following:

• Operating system is functioning properly.

• Network protocols are functioning properly. Test by using the ping command.

• Domain name to IP address translation is correct. Test by using the

nslookup

command with the fully qualified domain name of the

NetWorker server, storage node, or client you are recovering.

• The operating system recognizes the necessary devices. If the devices are not recognized, you might need to:

– Load the SCSI driver.

– Install the device driver software.

4. To complete the recovery process:

For a NetWorker server, see "Recovering a NetWorker Server" on page 71 .

70

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

• For a NetWorker client or storage node, see

"Recovering a NetWorker

Client or Storage Node" on page 87 .

Recovering a NetWorker Server

This section explains how to recover a NetWorker server on a Windows 2003 or Windows 2000 host computer.

Important:

Before you begin a NetWorker server recovery, you should understand the information about SYSTEM or VSS SYSTEM save sets in the

NetWorker Administrator’s Guide.

NetWorker Server Recovery Requirements

To recover a NetWorker server requires:

• The same version and patch level of the NetWorker software that was in use immediately prior to the disaster.

• The installation path of the NetWorker software prior to the disaster.

• Backup or clone volumes that contain the NetWorker server’s most recent bootstrap, client file indexes, and file-system data including all Windows

SYSTEM or VSS SYSTEM save sets.

If you routinely move NetWorker backup media to an offsite location for safekeeping, ensure that all necessary volumes are available so that you can avoid delays during a recovery. To list the media associated with the files you want to recover, run mminfo -mv from the command prompt. For more information about the mminfo command, refer to the NetWorker Command

Reference Guide.

How to Recover a Windows NetWorker Server

To recover a Windows NetWorker server, perform the following tasks in the order specified:

"Task 1: Reinstall the NetWorker Server Software" on page 72

"Task 2: Configure the NetWorker Server" on page 73

"Task 3: Locate the NetWorker Server Bootstrap Save Set ID" on page 75

"Task 4: Recover the NetWorker Server Bootstrap" on page 77

NetWorker V7.3

Disaster Recovery Guide

71

4

4

Chapter 4: Windows Disaster Recovery

"Task 5: Rename the Resource Files" on page 79

"Task 6: Reset and Inventory the Autochanger" on page 79

"Task 7: Create a Record of Backups Done After Bootstrap Created" on page 81

"Task 8: Recover the NetWorker Server’s Client File Indexes" on page 81

"Task 9: Recover the NetWorker Server Data" on page 83

"Task 10: Recover the Removable Storage Jukebox Configuration" on page 85

"Task 11: Verify the NetWorker Server Recovery" on page 86

Important:

When recovering a NetWorker server that was also being used as a EMC License Manager server (not recommended), the lictype.res file of the

EMC License Manager is not recovered during the NetWorker server recovery.

This is because the EMC License Manager files and directories are not considered part of the NetWorker software. After completing the disaster recovery on the NetWorker server, you must explicitly recover the EMC

License Manager as a client of the NetWorker server.

72

Task 1: Reinstall the NetWorker Server Software

Important:

If you updated from NetWorker V6.x and have not yet performed a scheduled backup of the NetWorker server, reinstall NetWorker V6.x and use it to recover the bootstrap, then update the NetWorker software and proceed with the recovery.

To reinstall the NetWorker server software:

1. Reinstall, to its original path location, the version of the NetWorker server software that was in use prior to the disaster. For instructions, refer to the

EMC NetWorker Installation Guide, Microsoft Windows Version.

During the installation, keep the following points in mind:

• If the NetWorker installation kit is available on a shared drive, you can reinstall it over the network.

• To update the NetWorker software, first recover the server to its state immediately prior to the disaster, then perform the update.

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

• You do not need to reload the license enablers; they will be recovered later in the process.

• Because the NetWorker mmrecov command is case sensitive, the

NetWorker installation path must be in the same case as the original.

For example, C:\Program Files\Legato\nsr is not same as C:\Program

Files\legato\NSR.

2. Reinstall any NetWorker patches that were in use prior to the disaster.

Task 2: Configure the NetWorker Server

To configure the NetWorker server:

1. In the NetWorker Administrator program, open the NetWorker server’s

Client resource and verify that the Aliases attribute (in the Preferences tab) contains the correct information. For example, aliases for a computer named kingdome might be: kingdome kingdome.seattle.washington.com

2. Create and configure the NetWorker server’s Device resources. Do not mount or relabel any volumes in the devices. For more information, refer to the EMC NetWorker Installation Guide, Microsoft Windows Version.

4

NetWorker V7.3

Disaster Recovery Guide

73

4

Chapter 4: Windows Disaster Recovery

If you are using an autochanger with Removable Storage disabled: a. Start a Windows command prompt and change to the

<NetWorker_install_path>\bin directory.

b. Create and configure the Autochanger resource using the NetWorker

Console Administration interface. For details about the NetWorker

Console Administration interface, refer to the NetWorker

Administrator’s Guide.

c. Run the nsrjb -vHE command from the command prompt. This resets the autochanger, ejects backup volumes, reinitializes the element status, and checks each slot for a volume. If the autochanger does not support the -E option (to reinitialize the element status), use the sjiielm command to initialize the element status. d. If you need to determine which volume contains the bootstrap, inventory the autochanger by running the nsrjb -Iv command from the command prompt. If you know the slot number where the bootstrap is located, use the nsrjb -Iv -S # command to inventory that particular slot.

e. If the device the NetWorker server is to use is connected to a remote storage node, the storage node device needs to be configured. For details, refer to the EMC NetWorker Installation Guide, Microsoft

Windows Version.

If you are using an autochanger with Removable Storage enabled: a. On the desktop, right-click My Computer and select Manage.

b. In the left pane of the Computer Management window, expand

Storage\Removable Storage\Physical Locations.

c. Right-click the icon for the autochanger and select Inventory. d. Minimize (but do not close) the Computer Management window.

e. After the Removable Storage inventory is complete, start the

NetWorker Administrator program and select Devices on the

Configure tab.

f. In the Devices window, right-click the Devices icon and select Create. g. In the Create Devices dialog box, create a stand-alone NetWorker

Device resource for each drive in the autochanger.

h. In the Computer Management window, double-click the icon for the autochanger (under Storage\Removable Storage\Physical Locations).

Icons for the autochanger’s volumes appear in the right pane.

i. Right-click the icon for the volume that contains the bootstrap and select Mount.

74

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

3. In the NetWorker Administrator program, configure the Client resource for the NetWorker server. Set the browse and retention policies to a time value that covers the oldest save sets being used for the recovery. This allows you to recover all of the NetWorker server’s records with the

mmrecov

command.

Important:

If the browse and retention policies are not long enough to cover all of the save sets you use, all of the NetWorker server’s records are recovered.

However, any records that are older than a month are discarded because the default browse policy is one month.

4. If the NetWorker server’s client file index has not been relocated from its original path, or if you are going to recover a relocated client file index from a backup created with NetWorker V7.x, skip this step and proceed to

"Task 3: Locate the NetWorker Server Bootstrap Save Set ID" on page 75 .

If you are going to recover the client file index from a backup created with a NetWorker version earlier than 7.x, and you moved the NetWorker server’s client file index path to a different location, you must: a. Edit the Index Path attribute of the NetWorker server’s Client resource to reflect the correct path.

b. Stop and restart the following NetWorker services:

– NetWorker Backup and Recover Server

– NetWorker Remote Exec Service

Task 3: Locate the NetWorker Server Bootstrap Save Set ID

If you already know the save set ID of the NetWorker server’s most recent bootstrap, skip this task and go to

"Task 4: Recover the NetWorker Server

Bootstrap" on page 77

.

The bootstrap contains the NetWorker server’s media database and resource configuration files, which are required for a NetWorker server recovery. For more information about the bootstrap, see

"Backing Up Data" on page 17

.

There are two ways to determine the save set ID of the most recent bootstrap:

• The faster way to determine the bootstrap save set ID is to find it in the

<NetWorker_install_path>\logs\savegrp.log file.

4

NetWorker V7.3

Disaster Recovery Guide

75

4

Chapter 4: Windows Disaster Recovery

76

By default, the NetWorker server automatically creates a bootstrap save set after each active scheduled backup that includes the server itself as a client. NetWorker software sends a copy of the bootstrap information to the default printer. You can also configure the NetWorker server to send an email of the bootstrap information after each scheduled backup. The

NetWorker server also writes the bootstrap information to the end of the

savegrp.log file that is written for the backup.

The following excerpt from a savegrp.log file shows a bootstrap save set ID of 1467388673 located on volume lacey.001. Also note the file and record numbers; having this information speeds the recovery process.

Jul 19 17:09:14 lacey: * lacey:bootstrap date time level ssid file record volume

07/19/01 05:08:50 PM full 1467388673 5 0 lacey.001

• If the bootstrap information is not available from the sources described previously, you can locate it by using the NetWorker scanner command, as follows, although this can be time-consuming: a. Insert the backup or clone volumes from the most recent scheduled backups into the appropriate device. (Do not use the NetWorker software to mount the media; the NetWorker server does not currently have the necessary information to recognize the media.) b. Start a Windows command prompt and change to the

<NetWorker_install_path>\bin directory.

c. If you are using an autochanger, load the first volume of the bootstrap save set into the first drive of the autochanger by running:

nsrjb -lnv -S

#

-f

device_name

where:

# is the slot where the first volume is located

device_name is the pathname for the first drive

You can obtain the device_name by running the NetWorker inquire command.

d. To determine the save set ID of the most recent bootstrap use the following command:

scanner -B

device_name

where device_name is the pathname for the drive where the appropriate volume is located (for example, \\.\Tape0).

If you do not locate the save set ID of the bootstrap from the most recent scheduled backup, run the scanner -B command on the volumes from the next most recent scheduled backup to locate the save set ID of the next most recent bootstrap.

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Task 4: Recover the NetWorker Server Bootstrap

In this task, you will recover the bootstrap save set. Recovering the bootstrap overwrites the media database in the <NetWorker_install_path>\mm directory and recovers the resource files to a temporary directory named

<NetWorker_install_path>\res.R.

Note:

The mmrecov command is used to recover only the media database and resource files. To recover the client file indexes (including the NetWorker server’s own client file index), use the nsrck command. For details about the

mmrecov

and nsrck commands, refer to the NetWorker Command Reference

Guide.

To recover the bootstrap save set:

1. Load the backup or clone volumes that contain the most recent bootstrap

save set. If you ran the scanner command in "Task 3: Locate the NetWorker

Server Bootstrap Save Set ID" on page 75 , you have already completed this

step; proceed to

step 2 .

If you did not run the scanner command in "Task 3: Locate the NetWorker

Server Bootstrap Save Set ID" on page 75

, do the following: a. Insert into the storage device the backup or clone volumes that contain the most recent bootstrap save set. (Do not use the NetWorker software to mount the volume; the NetWorker server does not currently have the necessary information to recognize the media.) b. If you are using an autochanger, run the following command to load the first volume of the bootstrap save set into the first drive of the autochanger:

nsrjb -lnv -S

#

-f

device_name

where:

– # is the slot where the first volume is located

– device_name is the pathname for the first drive

Note:

You can obtain the device_name by using the inquire command.

2. To recover the NetWorker server’s bootstrap, run the following command:

mmrecov

3. If you are using clone volumes and have not loaded all of the necessary volumes, mmrecov prompts you for the original volume. In that case: a. Press [Ctrl]+[c] to exit mmrecov.

b. Delete the records of the original volumes by running the following command:

NetWorker V7.3

Disaster Recovery Guide

77

4

4

Chapter 4: Windows Disaster Recovery

nsrmm -d

volume_name

For example, if the NetWorker software requests the original volume

mars.1 but that volume is not available, run the following command to delete mars.1 from the media database:

nsrmm -d mars.1

c. Restart the following NetWorker services:

– NetWorker Backup and Recover Server

– NetWorker Remote Exec Service d. Run the mmrecov command to recover the bootstrap.

4. If the server has multiple devices configured and enabled, the following message appears:

What is the name of the device you plan on using

[\\.\Tape1]?

If you receive this message, enter the name of the device you are using for the recovery. For example, enter

\\.\Tape0

5. When the following message appears, enter the save set ID for the latest bootstrap. If you are recovering the bootstrap from a clone volume, enter the save set ID associated with the clone.

Enter the latest bootstrap save set ID: 20076

6. When the following message appears, enter the file number to begin the recovery (if known) or accept the default of zero.

Enter starting file number (if known) [0]: 130

7. When the following message appears, enter the first record number to begin the recovery (if known) or accept the default of zero.

Enter starting record number (if known) [0]: 0

8. When the following message appears, ensure that the volume that contains the correct save set is inserted into the correct backup device, then press [Enter]:

Please insert the volume on which save set ID 20076 started into \\.\Tape0. When you have done this, press [Enter]:

9. Once the appropriate volume is loaded, the following message appears:

Scanning \\.\Tape0 for save set 20076; this may take a while...

78

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

The NetWorker software scans the volume for the appropriate save set and then begins the recovery by using the following commands:

• The nsrmmdbasm command is used to recover the media database.

• The uasm command is used to recover the resource files.

Use the NetWorker Administrator program to monitor the recovery.

Note:

The media database may not be on the same volume as the resource files. In this case, the recovery prompts you when additional media is required.

10. If the NetWorker Console server was installed on the same machine as the

NetWorker server, re-install the console software. For installation instructions, refer to the appropriate EMC NetWorker Installation Guide.

After reinstalling the NetWorker Console server, add the NetWorker server to list of hosts managed by the NetWorker Console server. For instructions on adding the server to the list of hosts, see the NetWorker

Administrator’s Guide.

Task 5: Rename the Resource Files

Because the NetWorker server’s resource files cannot reliably be overwritten while the NetWorker services are running, the mmrecov command recovers the res directory as res.R.

To put the recovered resource files into effect:

1. Stop the following NetWorker services on the NetWorker server:

• NetWorker Backup and Recover Server

• NetWorker Remote Exec Service

2. Copy the contents of the <NetWorker_install_path>\res.R directory to the

<NetWorker_install_path>\res directory.

3. Restart the NetWorker services.

Task 6: Reset and Inventory the Autochanger

If you are using a stand-alone storage device to perform the recovery, skip this

task and proceed to "Task 7: Create a Record of Backups Done After Bootstrap

Created" on page 81

.

4

NetWorker V7.3

Disaster Recovery Guide

79

4

Chapter 4: Windows Disaster Recovery

How to Reset and Inventory an Autochanger with Removable

Storage Disabled

To reset and inventory an autochanger with Removable Storage disabled:

1. Start a Windows command prompt and change to the directory that contains the NetWorker binary files. The default location is

<NetWorker_install_path>\bin.

2. To reset the autochanger, eject volumes, reinitialize the element status, and check each slot for a volume, run the following command:

nsrjb -vHE

3. If you know which slots have changed configuration since the disaster, inventory only those slots. For example, to inventory only slot 3, run the following command:

nsrjb -Iv -S 3

If you do not know which slots have changed configuration since the disaster, inventory all slots by running the following command:

nsrjb -Iv

Note:

If the autochanger does not support the -E option of the nsrjb command

(to reinitialize the element status) use the sjiielm command to initialize the element status. For more information about the nsrjb and sjiielm commands, refer to the NetWorker Command Reference Guide.

How to Reset and Inventory an Autochanger with Removable

Storage Enabled

To reset and inventory an autochanger with Removable Storage enabled:

1. In the NetWorker Administrator program, select Autochanger from the

Configure tab.

2. In the Autochanger window, right-click the icon for the Removable

Storage jukebox and select Delete.

3. Start a Windows command prompt and change to the directory that contains the NetWorker binary files. The default location is

<NetWorker_install_path>\bin.

4. Create a new Autochanger resource using the NetWorker Console

Administration interface. For details about the NetWorker Console

Administration interface, refer to the NetWorker Administrator’s Guide.

80

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

5. In the NetWorker Console Administration Libraries screen, right-click the icon for the Removable Storage jukebox and select Operations.

6. Right-click My Computer and select Manage.

7. In the left pane of the Computer Management window, expand

Storage\Removable Storage\Media Pools\Import and check the appropriate type-specific media pool for NetWorker volumes.

8. On the Autochanger Operations dialog box, click Add Volume to reallocate NetWorker volumes from the import pool to this jukebox.

Continue allocating until all of the NetWorker volumes belonging to the library associated with this jukebox are moved from the Import pool to the appropriate type-specific pool under Storage\Removable Storage\Media

Pools\Legato NetWorker.

9. In the Autochanger Operations dialog box, click the Reset button to reset the Removable Storage jukebox.

10. In the Autochanger Operations dialog box, click the Inventory button to inventory the contents of the Removable Storage jukebox.

Task 7: Create a Record of Backups Done After Bootstrap Created

After a disaster recovery, the media database will not contain information about any backups done after the bootstrap was created. The media database retains information about the last known file/record marks on the tape, and if backups were done after the bootstrap was created, the file/record information will not be current. For any tapes that contain backups that were completed after the bootstrap was created, use scanner with, at a minimum, the -m option to scan the save sets and/or tapes into the media database. This will ensure that the media database has a record of all completed savesets and that valid backups will not be overwritten by future backups.

Additionally, write protect all appendable volumes until you can complete the scanner activity to prevent scheduled backups from using these volumes. See

"Task 8: Create a Record of Backups Done after Bootstrap Created" on page 37

for instructions on write protecting and scanning.

Task 8: Recover the NetWorker Server’s Client File Indexes

After the NetWorker server’s bootstrap save set is recovered, you can recover the client file indexes in any order; it is not necessary to recover the server’s own client file index before recovering the index of any other client.

4

NetWorker V7.3

Disaster Recovery Guide

81

4

Chapter 4: Windows Disaster Recovery

Important:

If the clients have the NetWorker client software installed, you can run manual and scheduled backups as soon as the NetWorker server bootstrap is recovered. You can also recover complete save sets. You cannot, however, browse a client’s save sets until you recover the client file index.

To recover the client file indexes:

1. To recover the client file indexes, run one of the following commands from the NetWorker server:

• For all clients:

nsrck -L7

• For a specific client:

nsrck -L7

client_name

2. If you are recovering the client file indexes from a clone volume, you may be prompted to load the original volume. In that case: a. Press [Ctrl]+[c] to exit nsrck.

b. Delete the records of the original volumes by running the following command:

nsrmm -d

volume_name

For example, if the NetWorker software requests the original volume

mars.1 but that volume is not available, run the following command to delete mars.1 from the media database:

nsrmm -d mars.1

c. Restart the following NetWorker services:

– NetWorker Backup and Recover Server

– NetWorker Remote Exec Service d. Recover the client file indexes by running one of the following commands from the command prompt:

– For all clients:

nsrck -L7

– For a specific client:

nsrck -L7

client_name

82

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Task 9: Recover the NetWorker Server Data

Important:

Recovery of volume mount points and their data requires special handling. If you attempt to recover a mount point and the mounted volume’s data in a single operation, the data will be recovered to the root of the host volume and recovery of the mount point will fail. To successfully recover the mounted volume’s data, first manually re-create the mount point. Use the same path as the original. Then, perform a separate NetWorker file recovery to recover just the mounted volume’s data (without including any of the host volume’s data in the recovery).

For more information about backing up, recovering, and creating mount points, refer to the NetWorker Administrator’s Guide.

Note:

This special handling is not required when performing an ASR recovery.

To recover the remainder of the NetWorker server data:

1. Log in to Windows with local administrator privileges.

Important:

Directed recovery of SYSTEM or VSS SYSTEM save sets is not supported. To recover these save sets, log into the computer being recovered with local administrator privileges. Additionally, recover the SYSTEM or VSS

SYSTEM save sets to the location that they were in before the disaster.

2. If you are using an autochanger, make sure it has been inventoried. See

"Task 6: Reset and Inventory the Autochanger" on page 79 . This ensures

that the NetWorker server can recognize the location of each volume.

3. If you load a clone volume, the NetWorker software uses the clone volume for the remainder of the recovery process if either of the following is true:

• The original volume was not added back into the media database.

• The original volume was added back into the media database, but it was not placed in an autochanger and inventoried.

However, if the original volume was added back into the media database, placed in an autochanger, and inventoried, the NetWorker software prompts you to mount the original volume.

4

NetWorker V7.3

Disaster Recovery Guide

83

4

Chapter 4: Windows Disaster Recovery

4. Ensure that all drives that contained system information prior to the disaster are accessible. The system information includes the boot partition, system partition, and the partition that contains the databases and database logs for the system state components.

5. Start the NetWorker User program.

6. Click the Recover button.

7. In the Source Client dialog box, select the computer you are recovering and click OK.

8. In the Destination Client dialog box, select the computer you are recovering and click OK.

9. The computer’s directory structure appears in the Recover window.

10. Mark the following save sets for recovery.

• All local drives (except the NetWorker installation directory)

• SYSTEM FILES:

• SYSTEM DB:

• SYSTEM STATE:

Alternatively, if you are using Windows 2003 and Volume Shadow Copy

Service (VSS) is licensed and enabled, mark the following save sets:

• All local drives (except the NetWorker installation directory)

• VSS SYSTEM BOOT:

• VSS SYSTEM FILESET:

• VSS SYSTEM SERVICES:

• VSS USER DATA:

• VSS OTHER:

• VSS ASR DISK:

Note:

If you recover save sets in multiple NetWorker User operations, recover the SYSTEM STATE or VSS SYSTEM BOOT save set last to avoid booting Windows twice.

NetWorker software backs up user profiles in the save set that corresponds to the disk drive on which they are located (for example, the C: save set).

To recover user profiles, recover the save set in which they were saved.

User profile folders are located in %SystemDrive%\Documents and Settings.

11. From the Option menu, select Recover Options.

84

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

12. In the Recover Options dialog box, select Overwrite Existing File.

If you do not select Overwrite Existing File, the recovery process pauses before any existing files are overwritten. A Naming Conflict dialog box appears, prompting you to indicate how the NetWorker software should resolve filename conflicts.

13. Click Start to begin the recovery.

Note:

Files in the Windows Recycle Bin are not password protected. If the

Password Protection dialog box appears, you can select Recover > OK to recover these files.

14. Check the log file to verify that no error messages were generated during the recovery sessions. The NetWorker software logs information about the recovery process to <NetWorker_install_path>\logs.

The log is overwritten each time a recovery is performed. If there are error messages in the log file, you might need to run the recovery again after addressing the source of the errors.

15. After recovery, reboot the computer to completely recover the system-protected files and restart services that were automatically stopped prior to recovery.

16. If the NetWorker Management Console server was installed on the host and the Console database was lost, recover the database. For instructions, see

"How to Recover the NetWorker Management Console Server

Database" on page 95 .

Note:

By default, the directed recover option is enabled when you install the

NetWorker client. If the directed recover option was disabled on the client before the disaster, you need to explicitly set this option to disable directed recoveries to this client. This option is set through the nsradmin command. For more information about setting the disable directed recover option on the

NetWorker client, refer to the NetWorker Administrator’s Guide.

Task 10: Recover the Removable Storage Jukebox Configuration

If you are using a stand-alone storage device or an autochanger with

Removable Storage disabled, skip this task and go to

"Task 11: Verify the

NetWorker Server Recovery" on page 86 .

To recover an autochanger with Removable Storage enabled:

1. Right-click My Computer and select Manage.

2. In the left pane of the Computer Management window, expand

Storage\Removable Storage\Physical Locations.

NetWorker V7.3

Disaster Recovery Guide

85

4

4

Chapter 4: Windows Disaster Recovery

3. Double-click the icon for your autochanger. Icons for the autochanger’s volumes appear in the right pane.

4. Right-click the icon for the volume that contains the bootstrap and select

Mount.

5. Start a Windows command prompt and change to the

<NetWorker_install_path>\bin directory.

6. From the command prompt, run the following command to recover the

NetWorker server’s bootstrap:

mmrecov

7. Stop the following NetWorker services:

• NetWorker Backup and Recover Server

• NetWorker Remote Exec Service

8. Rename the <NetWorker_install_path>\res.R directory to

<NetWorker_install_path>\res.

9. Restart the NetWorker services.

10. On the Autochanger Operations dialog box, click Reset.

Task 11: Verify the NetWorker Server Recovery

To verify the NetWorker server recovery:

1. Reboot the NetWorker server host computer and verify that the

NetWorker Backup and Recover Server, NetWorker Power Monitor, and

NetWorker Remote Exec services are started.

2. Use the Windows Event Viewer to examine the event logs for errors. In particular, check for:

• Service startup errors related to the Windows system state

• Errors regarding the recovery of Windows system-protected files

For information about how the NetWorker software handles the Windows system state and system-protected files, refer to the NetWorker

Administrator’s Guide.

3. Verify that the NetWorker server and its associated clients are included in a scheduled NetWorker backup.

4. Perform a test backup and recover.

86

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

5. Verify that any applications that were running prior to the disaster, such as

Microsoft Office, have been properly recovered. To check this, you can start each application and open a file.

6. If the NetWorker server was previously configured as a domain controller, verify that the drives configured to store the Active Directory database and log files have been recovered.

Recovering a NetWorker Client or Storage Node

This section explains how to recover a Windows NetWorker client host computer or a Windows 2003 or Windows 2000 storage node host computer.

Notes

:

• For Windows 2003 and Windows XP Professional systems, perform tasks

1 through 4 that appear on the following pages — but do so only as a last resort for recovering a NetWorker client. For a NetWorker client, use the

ASR recovery method if possible. For more information, see

"Performing a Windows ASR Recovery" on page 55 . You cannot use the ASR recovery

method for a NetWorker storage node.

• Before you begin the recovery procedures, identify all EMC products

(such as SmartMedia

®

, ClientPak

®

, and NetWorker Modules) and any associated patches that were installed prior to the disaster. You must reinstall any EMC backup software and patches that are unusable or appear to be damaged. For more information, refer to the appropriate product installation guide.

• You can recover a NetWorker client or storage node to the original computer, or to a different computer.

To recover the client or storage node, perform the following tasks in the specified order:

"Task 1: Satisfy the NetWorker Client or Storage Node Recovery

Requirements" on page 88 .

"Task 2: Reinstall the NetWorker Client or Storage Node Software" on page 89 .

"Task 3: Recover the NetWorker Client or Storage Node Data" on page 89

"Task 4: Verify the NetWorker Client or Storage Node Recovery" on page 92

Note:

Before you begin recovery of a NetWorker client or storage node, you should understand the information about SYSTEM or VSS SYSTEM save sets in the NetWorker Administrator’s Guide.

NetWorker V7.3

Disaster Recovery Guide

87

4

4

Chapter 4: Windows Disaster Recovery

Partial Recovery of a NetWorker Storage Node

In most cases, the NetWorker server should be fully functional before you begin recovery of its associated clients or storage nodes. However, if the storage device required for a NetWorker server recovery is connected to a remote storage node, you might have to partially recover the storage node

before you can recover the NetWorker server. In that case:

• Reinstall the NetWorker client or storage node software to provide a storage device for the NetWorker server to use. For information, see

"Task

2: Reinstall the NetWorker Client or Storage Node Software" on page 89 .

• Recover the NetWorker server. For information, see

"Recovering a

NetWorker Server" on page 71 .

• Perform the remaining tasks to recover the client or storage node. For more information, see

"Task 3: Recover the NetWorker Client or Storage

Node Data" on page 89

and

"Task 4: Verify the NetWorker Client or

Storage Node Recovery" on page 92 .

Task 1: Satisfy the NetWorker Client or Storage Node Recovery

Requirements

To recover a NetWorker client or storage node requires:

• The same version and patch level of the NetWorker software that was in use prior to the disaster.

• The installation path of the NetWorker software prior to the disaster.

• Backup or clone volumes that contain the NetWorker client or storage node data.

Note:

If you routinely move NetWorker backup media to an offsite location for safekeeping, ensure that you have all necessary volumes available to avoid delays during a recovery. You can run the mminfo -s server -mv command from the command prompt to list the media associated with the files you intend to recover. For details about the mminfo command, refer to the

NetWorker Command Reference Guide.

88

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Task 2: Reinstall the NetWorker Client or Storage Node Software

If the NetWorker client or storage node software is not properly installed and running on the host computer, you must reinstall it, as follows:

1. To reinstall the NetWorker client or storage node software, refer to the appropriate EMC NetWorker Installation Guide.

• If the NetWorker installation kit is available on a shared drive, reinstall it over the network.

• Reinstall the NetWorker software to the same path location it occupied prior to the disaster.

• To update the NetWorker software, first recover the NetWorker client or storage node to its predisaster state, then perform the update.

2. Reinstall any NetWorker patches that were installed prior to the disaster.

3. Reinstall the temporary enabler code for each EMC backup product. For temporary enabler codes, refer to the product installation guide.

4. If you are recovering a NetWorker storage node, ensure that the storage devices the NetWorker server will use are correctly configured. For details, refer to the appropriate NetWorker Administrator’s Guide.

5. From the NetWorker User program, perform a test recovery to verify that the NetWorker recovery process is functioning properly.

Task 3: Recover the NetWorker Client or Storage Node Data

Important:

Recovery of volume mount points and their data requires special handling. If you attempt to recover a mount point and the mounted volume’s data in a single operation, the data will be recovered to the root of the host volume and recovery of the mount point will fail. To successfully recover the mounted volume’s data, first manually re-create the mount point. Use the same path as the original. Then perform a separate NetWorker file recovery to recover just the mounted volume’s data (without including any of the host volume’s data in the recovery).

For more information about backing up, recovering, and creating mount points, refer to the NetWorker Administrator’s Guide.

Note:

This special handling is not required when performing an ASR recovery.

4

NetWorker V7.3

Disaster Recovery Guide

89

4

Chapter 4: Windows Disaster Recovery

To recover the NetWorker client or storage node data:

1. Log on to Windows with local administrator privileges.

90

Important:

Directed recovery of SYSTEM or VSS SYSTEM save sets is not supported. To recover these save sets, log onto the computer being recovered with local administrator privileges. Additionally, recover the SYSTEM or VSS

SYSTEM save sets to the location that they were in before the disaster.

2. Ensure that all drives that contained system information prior to the disaster are accessible. The system information includes the boot partition, system partition, and the partition that contains the databases and database logs for the system state components.

3. Start the NetWorker User program.

4. Click the Recover button.

5. In the Source Client dialog box, select the computer you are recovering and click OK.

6. In the Destination Client dialog box, select the computer you are recovering and click OK.

7. The computer’s directory structure appears in the Recover window. Mark all appropriate save sets for recovery:

Windows 2003 and Windows XP Professional

:

• All local drives (except the NetWorker installation directory)

• SYSTEM STATE:

• SYSTEM FILES:

• SYSTEM DB:

Alternatively, if you are using Windows 2003 and Volume Shadow Copy

Service (VSS) is licensed and enabled, mark the following save sets:

• All local drives (except the NetWorker installation directory)

• VSS SYSTEM BOOT:

• VSS SYSTEM FILESET:

• VSS SYSTEM SERVICES:

• VSS USER DATA:

• VSS OTHER:

• VSS ASR DISK:

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

Windows 2000

:

• All local drives (except the NetWorker installation directory)

• SYSTEM STATE:

• SYSTEM FILES:

• SYSTEM DB:

• SHAREPOINT: (for SharePoint Portal Servers only)

Windows NT 4.0

:

• All local drives (except the NetWorker installation directory)

• SYSTEM STATE:

• REPAIRDISK: (recover only if needed)

Notes

• If you recover save sets in multiple NetWorker User operations, recover the SYSTEM STATE or VSS SYSTEM BOOT save set last to avoid booting

Windows twice.

• NetWorker software backs up user profiles in the save set that corresponds to the disk drive on which they are located (for example, the

C: save set). To recover user profiles, recover the save set in which they were saved. User profile folders are located in:

– Windows 2003, Windows 2000, and Windows XP Professional:

%SystemDrive%\Documents and Settings

– Windows NT 4.0:

%SystemRoot%\profiles

8. From the Option menu, select Recover Options.

9. In the Recover Options dialog box, select Overwrite Existing File.

If you do not select Overwrite Existing File, the recovery process pauses before any existing files are overwritten. A Naming Conflict dialog box appears, prompting you to indicate how the NetWorker software should resolve filename conflicts.

10. Click Start to begin the recovery.

Note:

Files in the Windows Recycle Bin are not password protected. If the

Password Protection dialog box appears, select Recover > OK to recover them.

11. Check the log file to verify that no error messages were generated during the recovery sessions. The NetWorker software logs information about the recovery process to the <NetWorker_install_path>\logs file. This file is

NetWorker V7.3

Disaster Recovery Guide

91

4

4

Chapter 4: Windows Disaster Recovery overwritten each time a recovery is performed. If there are error messages in the log file, you might need to run the recovery again after addressing the source of the errors.

12. After recovery, reboot the computer to completely recover the system-protected files and restart services that were automatically stopped prior to recovery.

Notes:

• If you perform a complete reinstallation of the Windows operating system during a disaster recovery, the display resolution will be set to default values when the disaster recovery is complete. To reconfigure to the previous display settings, use the Control Panel Display tool. This problem does not occur if you perform an ASR recovery on a Windows

2003 or Windows XP Professional NetWorker client.

• By default, the directed recover option is enabled when you install the

NetWorker client. If the directed recover option was disabled on the client before the disaster, you need to explicitly set this option to disable directed recoveries to this client. This option is set through the nsradmin command. For more information about setting the disable directed

recover

option on the NetWorker client, refer to the NetWorker

Administrator’s Guide.

Task 4: Verify the NetWorker Client or Storage Node Recovery

To verify the NetWorker client or storage node recovery:

1. Reboot the NetWorker client or storage node host computer and verify that the NetWorker Remote Exec service has started.

2. Use the Windows Event Viewer to examine the event logs for errors. In particular, check for:

• Service startup errors related to the Windows system state

• Errors regarding the recovery of Windows system-protected files

For information about how the NetWorker software handles the Windows system state and system-protected files, refer to the NetWorker

Administrator’s Guide.

3. Ensure that the storage node and the NetWorker server can detect the storage devices the NetWorker server is to use.

4. Perform a test recovery by using each of the EMC backup products you have reinstalled. Recover data with the EMC product used to back it up.

Refer to the product documentation as necessary.

92

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery

5. Perform a test backup to the NetWorker server to ensure that the connection between the NetWorker client or storage node and the

NetWorker server is working properly.

6. Verify that any applications (such as Microsoft Office) that were running prior to the disaster have been properly recovered. To verify this, run each application and open a previously saved document.

7. If you want to use Active Desktop, install it after you have recovered all of the computer’s data.

8. If the NetWorker client or storage node computer was previously configured as a domain controller:

Windows 2000

— Verify that the drives configured to store the Active

Directory database and log files have been recovered.

Windows NT 4.0

— If the computer is a Primary Domain Controller, synchronize it with the domain. If the computer is a Backup Domain

Controller, synchronize it with the Primary Domain Controller.

Recovery of DHCP and WINS Databases

If you are using Windows 2003 and are using ASR backup and recovery or have VSS licensed and enabled, no special handling is required to back up and recover DHCP and WINS databases. However, if you are not using ASR backup and recovery or do not have VSS licensed and enabled, use this section to learn how to back up and recover the DHCP and WINS databases.

Important:

To recover the DHCP database, you must have a NetWorker backup that includes the %SystemRoot%\System32\dhcp directory. To recover the WINS database, you must have a NetWorker backup that includes a local disk backup of the WINS database. For instructions on configuring DHCP and

WINS servers to include these databases in scheduled NetWorker backups, refer to the NetWorker Administrator’s Guide.

4

NetWorker V7.3

Disaster Recovery Guide

93

4

Chapter 4: Windows Disaster Recovery

How to Recover a DHCP Database

To recover a DHCP database:

1. Use the NetWorker User program to recover the backup of the

%SystemRoot%\System32\dhcp directory on the DHCP server.

2. Use the Microsoft DHCP administrative tools to recover the DHCP database. For detailed instructions, refer the DHCP database recover procedures in the Microsoft documentation.

How to Recover a WINS Database

To recover a WINS database:

1. Use the NetWorker User program to recover the backup of the WINS database to a drive on the WINS server.

2. Use the Microsoft WINS administrative tools to recover the WINS database. For detailed instructions on using Microsoft WINS administrative tools, refer to the Microsoft documentation.

Recovering the NetWorker Management Console Server

To recover from the NetWorker Management Console server after a disaster, perform three steps:

1. Reinstall NetWorker software, or recover the software using the disaster recovery procedures described earlier in this chapter.

2. Reinstall Console server software, or recover the software using normal recovery procedures as described in the NetWorker Administrator’s Guide.

3. Recover the NetWorker Management Console server database.

During recovery of the NetWorker Management Console server database, no

NetWorker Console graphical interface will be available. As a result, any messages such as mount requests will not be able to be addressed from the

Console.

To resolve this issue, temporarily install the NetWorker Management Console server on another machine. Add the appropriate NetWorker server to the

Enterprise screen and set permissions appropriately so the Console user can administer the NetWorker server.

94

NetWorker V7.3

Disaster Recovery Guide

Chapter 4: Windows Disaster Recovery use the nsrwatch command to view messages and the appropriate commands, such as nsrjb, to address the issues. For more information about nsrwatch, nsrjb, and other NetWorker commands, refer to the Command Reference Guide.

How to Recover the NetWorker Management Console Server Database

If the NetWorker Management Console server database becomes corrupted or if the Console server is damaged, use this procedure to recover lost data. When a disaster recovery is necessary, the NetWorker client software must be reinstalled or recovered before recovering the Console database.

For information about backing up the Console database, please see the

NetWorker Administrator’s Guide.

To recover the Console server database:

1. Stop the Console server if it is currently running.

2. At the command prompt, enter the recoverpsm command:

recoverpsm

recover options

Where recover options can be one of the following:

-s

— Specifies the name of the NetWorker server.

-c

— Specifies the name of the NetWorker client.

-f

— Overwrites the existing database file, if one is present.

-t

— Specifies the point in time of the backup that will be recovered.

If this is not specified, the most recent backup is used.

-d

— Specifies the destination directory where the database files will be recovered. If this is not specified, the database is restored to the current Console server database directory.

-S

— If the Console server has been moved to a different machine after the last backup, use this option to specify the name of the previous Console server host. The format of the database server name is

gst_on_

<NMC server host name>. <NMC server host name> should be the machines "short name;" for example, if the machine name is

wolf.legato.com, the proper format for this option would be

-S gst_on_wolf

.

3. Restart the Console server.

For more information about the recoverpsm command, see the NetWorker

Command Reference Guide.

4

NetWorker V7.3

Disaster Recovery Guide

95

4

Chapter 4: Windows Disaster Recovery

96

NetWorker V7.3

Disaster Recovery Guide

Chapter 5: Mac OS X Disaster Recovery

This chapter explains how to recover from disaster on a Mac OS X host running the NetWorker 7.3 client.

Recovering a NetWorker Client

The following sections provide information on recovering a NetWorker client:

"Prerequisites" on page 97

"How to Recover a NetWorker Client" on page 98

Prerequisites

Before recovering the NetWorker client, ensure that the Mac OS X operating system is installed on the computer and that the NetWorker server is functional and available on the network.

If you need to recover the NetWorker server, see one of the following chapters as appropriate:

"Chapter 3: UNIX Disaster Recovery" on page 21

"Chapter 4: Windows Disaster Recovery" on page 49

"Chapter 6: NetWare Disaster Recovery" on page 101

Additionally, ensure that you have the following:

• Version and patch level of the NetWorker client.

• The name of the computer on which the NetWorker server is running.

• A successful backup of the Open Directory databases. For more information about backing up the Open Directory databases, refer to the

NetWorker Administrator’s Guide.

NetWorker V7.3

Disaster Recovery Guide

97

5

Chapter 5: Mac OS X Disaster Recovery

How to Recover a NetWorker Client

This section describes how to recover a NetWorker client to the original computer or to a different computer of the same operating system.

To recover a NetWorker client, complete the following tasks:

"Task 1: Reinstall the NetWorker Client Software" on page 98

"Task 2: Recover the Application and User Data" on page 98

"Task 3: Restore Open Directory Database Files (Mac OS X Server)" on page 99

"Task 4: Perform a Test Backup and Recovery" on page 100

Task 1: Reinstall the NetWorker Client Software

To reinstall the NetWorker client software:

1. Reinstall the same version of the NetWorker client.

Note:

If you want to upgrade the client software, first recover the client to its original state, and then perform the upgrade.

2. Reinstall any NetWorker client patches that were installed before the disaster.

3. (Optional) Use the recover program to perform a test recovery to ensure that the recovery process is functioning properly.

Task 2: Recover the Application and User Data

To recover the application and user data:

1. If you need to determine which volumes contain the application and user data backups for this computer, use the mminfo -avot command on the

NetWorker server, for example:

$ mminfo -avot -c

client_name

where client_name is the hostname of the computer whose application and user data are being recovered.

2. Open a recover prompt on the NetWorker client with the following command:

$ recover

3. At the recover prompt, browse backed-up Mac OS X data by using traditional UNIX file-system navigation commands.

98

NetWorker V7.3

Disaster Recovery Guide

Chapter 5: Mac OS X Disaster Recovery

4. Add all the directories and files to be recovered, by entering the add command, for example: recover> add

directory_name

To automatically overwrite existing files, enter the force option with the

add

command.

5. Start the recovery by entering the following command at the recover prompt: recover> recover

Important:

When recovering application and user data, do not recover any

Mac OS X operating system boot files that should not be overwritten. For example, do not recover the Mac OS X operating system kernel, /mach_kernel.

Task 3: Restore Open Directory Database Files (Mac OS X Server)

This procedure assumes the NetWorker client was configured to backup Open

Directory files using a savepnpc script as described in NetWorker

Administrator’s Guide. You must be a NetWorker administrator to perform this procedure.

To restore Open Directory database files after a catastrophic failure requiring reinstallation of Mac OS X Server software:

1. Ensure the host is configured as a standalone server: a. Open Server Admin.

b. Select Open Directory and click Settings c. Click General, and change Role to Standalone Server.

2. Open a Terminal session as root.

3 . Using the recover program, restore Open Directory's exported databases and configuration files:

# recover -af /etc/openldap /var/backups/networker.odpdb

/etc/hostconfig /var/backups/networker.ldif

/var/backups/networker.nidump

4. Restore Open Directory's LDAP database using the following command:

# slapadd -c -l /var/backups/networker.ldif

NetWorker V7.3

Disaster Recovery Guide

99

5

5

Chapter 5: Mac OS X Disaster Recovery

5. If your LDAP server uses SSL, restore the Open Directory Password Server database using the following command:

# mkpassdb -mergedb /var/backups/networker.odpdb

6. Restore the local NetInfo domain using the following command:

# niload -r / . < /var/backups/networker.nidump

7. Use the recover program to restore Open Directory preferences:

# recover -af /Library/Preferences/DirectoryService

8. Start the LDAP server:

# /sbin/SystemStarter start LDAP

Task 4: Perform a Test Backup and Recovery

To test the NetWorker client backup and recovery process:

1. Perform a test manual backup by using the NetWorker backup commands.

2. Perform a test recovery by using the NetWorker recover commands.

For additional information on testing backup and recovery configurations, refer to the NetWorker Administrator’s Guide.

100

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

This chapter provides instructions on how to perform a disaster recovery on a

NetWare system that uses the NetWorker software.

Important:

This chapter applies only to NetWorker clients and servers running

NetWorker V4.2.x, NetWare version.

This chapter includes the following sections:

"Replacing a Hard Drive" on page 101

"Recovering the NetWare Operating System" on page 103

"Recovering a NetWare NetWorker Server" on page 107

"Recovering a Replicated NDS Partition" on page 114

Note:

All of the utilities referenced in this chapter are NetWare console utilities.

Replacing a Hard Drive

To replace a hard drive:

"Obtain the Hard Drive Information" on page 102

"Replace the Hard Drive" on page 102

"Completing the Recovery Process" on page 103

NetWorker V7.3

Disaster Recovery Guide

101

6

Chapter 6: NetWare Disaster Recovery

Obtain the Hard Drive Information

Before you replace the hard drive, you must obtain the following information about the drive:

• Size of the drive

• Size, format, and volume label assigned to each drive partition

• NetWare Directory Services (NDS), the NDS tree topology, and the location of server objects, partitions and replicas, and bindery context settings

You can obtain information about each drive by using:

install

(or nwconfig on NetWare 5.0). This server utility provides information about each hard drive including each volume segment on the drive.

netadmin

in DOS or the NetWorker Administrator program (nwadmin) in Windows.

dsrepair

. This server utility enables you to perform a check on each partition and replica.

autoexec.ncf. This file contains the bindery context settings.

ndir

. This workstation utility provides information about each volume, directory, and file on the disk.

Replace the Hard Drive

For detailed instructions on how to replace failed hard drives, refer to the appropriate NetWare documentation and the appropriate hard drive vendor documentation.

Important:

Install a new drive that is the same size or larger than the original drive. This will ensure that you can recover all of the drive’s data.

Do not delete any volume objects from the NDS tree. This would eliminate any references that other objects might have to a particular volume.

102

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

Completing the Recovery Process

To complete the recovery process after replacing the hard drive, see the following sections:

"Recovering the NetWare Operating System" on page 103

"Recovering a NetWare NetWorker Server" on page 107

"Recovering a Replicated NDS Partition" on page 114

Note:

To recover computer data by using NetWorker software, refer to the

NetWorker Administrator’s Guide.

Recovering the NetWare Operating System

The following sections contain information about recovering the NetWare operating system:

"Prerequisites" on page 103

"How to Recover the NetWare Operating System" on page 104

"Completing the Recovery Process" on page 106

Prerequisites

To recover the operating system, you need the following:

• NetWare operating system, license software, patch level, and version

• Version and patch level of the MS-DOS software

• IPX/SPX and/or TCP/IP properties:

– Adapter type

– IPX network number or IP address

– Default gateway

– Subnet mask

– DNS server

• Computer properties:

– Computer name

– DNS domain name

– Administrator name and password

• Device and SCSI drivers

• Startup files, including autoexec.bat, config.sys, autoexec.ncf, and startup.ncf

NetWorker V7.3

Disaster Recovery Guide

103

6

6

Chapter 6: NetWare Disaster Recovery

How to Recover the NetWare Operating System

This section describes how to recover the NetWare operating system by using the NetWorker software. Use this process to recover the NetWare operating system back to the original computer, or to a different computer with the same name.

You can recover the operating system by performing:

• A complete installation by using install or nwconfig. In this instance, you perform a complete reinstallation and configuration of the operating system.

• Alternatively, perform a partial installation by using install or nwconfig.

In this case, reinstall and configure only those files that enable the computer to communicate over the network. Then, recover the remaining operating system and configuration files by using the NetWorker software.

To recover the operating system, complete the following tasks:

"Task 1: Reinstall the Operating System" on page 104

"Task 2: Configure the Operating System" on page 105

"Task 3: Test the Operating System and Configure Devices" on page 106

Task 1: Reinstall the Operating System

To reinstall the operating system:

1. Reinstall MS-DOS, including the AUTOEXEC.BAT and CONFIG.SYS files, into its original location. For details, refer to the MS-DOS documentation.

2. Reinstall the same version and patch level of NetWare by using the install or nwconfig program. Reinstall NetWare into its original location with the same:

• Version

• Patch level

• Computer name

• IPX network number and/or IP address

For details, refer to the NetWare documentation.

If you want to upgrade NetWare, first recover the operating system to its original state, and then perform the upgrade.

104

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

3. Reinstall NDS into its original location by using the install or nwconfig utility:

• Use the same name as the original tree.

• Use the same server name.

• Use the same container and Administrator name.

• Ensure the Administrator object resides at the same level, in the same container, as it did in the original tree.

For details, refer to the NetWare documentation.

Important:

If this server will become the master of the NDS Root partition, during installation, make sure you re-create the Organization object by using the same name. If the Organization object does not have the same name, the subtree will contain new empty containers.

Task 2: Configure the Operating System

To configure the operating system:

1. Configure the IPX/SPX and TCP/IP properties exactly as they were configured before. If you do not, the NetWorker software will assign the computer a new host ID. If the computer has a different host ID, you must reregister the NetWorker software. For details, refer to the EMC NetWorker

Installation Guide, NetWare Version.

2. If you are recovering to a new computer, assign the same hostname for the new computer. If you do not use the same hostname, you will not be able to recover the NetWorker indexes associated with the original computer.

3. Configure the date and time properties as they were configured before.

4. If you had any additional NetWare components or services, reinstall them before recovering the computer’s data.

5. Restart the computer after installing the NetWare operating system.

6

NetWorker V7.3

Disaster Recovery Guide

105

6

Chapter 6: NetWare Disaster Recovery

Task 3: Test the Operating System and Configure Devices

To test the operating system and configure the devices:

1. Configure any devices the NetWorker software requires (for example,

SCSI pass-through devices for autochangers).

2. Perform a test to verify that the:

• Name-to-address resolution is correct. Test by running nslookup and by using the fully-qualified domain name of the NetWare NetWorker backup server, client, or storage node.

• Network protocols are functioning properly.

• Operating system is functioning properly.

• Time synchronization is functioning properly.

• The operating system recognizes the devices. If the operating system does not recognize the devices, you might need to:

– Modify the device configuration files to enable the computer to communicate with the device during recovery.

– Load the SCSI driver and its associated ASPI driver.

– Install the device driver.

Completing the Recovery Process

To complete the recovery process after reinstalling and configuring the operating system, use the following procedures, depending on the type of

NetWorker computer you are recovering:

"Recovering a NetWare NetWorker Server" on page 107

"Recovering a Replicated NDS Partition" on page 114

Note:

To recover computer data by using the NetWorker software, refer to the

NetWorker Administrator’s Guide.

106

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

Recovering a NetWare NetWorker Server

The following sections provide information on recovering a NetWorker server:

"Prerequisites" on page 107

"How to Recover a NetWare NetWorker Server" on page 107

Prerequisites

To recover the NetWorker server, you need the following:

• Original version and patch level of NetWorker software

• Original location of NetWorker software

• Backup or clone volumes containing the:

– NetWorker server media manager. By default, this directory is located in the SYS:NSR\MM directory.

– NetWorker server indexes. By default, this directory is located in the

SYS:NSR\INDEX\<server_name> directory.

– NetWorker server configuration files. By default, this directory is located in the SYS:NSR\RES directory.

How to Recover a NetWare NetWorker Server

To recover a NetWorker server, complete the following tasks:

"Task 1: Reinstall the NetWorker Software" on page 108

"Task 2: Configure NetWorker Device and Client Resource" on page 108

"Task 3: Locate the Server’s Bootstrap Save Set ID" on page 109

"Task 4: Recover the NetWorker Server Bootstrap" on page 109

"Task 5: Rename the NetWorker Server Resource Files" on page 111

"Task 6: Recover the NetWorker Server Data" on page 111

"Task 8: Recover NDS Data" on page 112

"Task 7: Create a Record of Backups Done after Bootstrap Created" on page 112

"Task 8: Recover NDS Data" on page 112

"Task 9: Complete the Server Recovery" on page 113

6

NetWorker V7.3

Disaster Recovery Guide

107

6

Chapter 6: NetWare Disaster Recovery

Task 1: Reinstall the NetWorker Software

To reinstall the NetWorker software:

1. Reinstall the same version of NetWorker software into its original location.

For installation instructions, refer to the EMC NetWorker Installation Guide,

NetWare Version.

• If you want to upgrade the NetWorker server, first recover the server to its original state, and then perform the upgrade.

• You do not need to reload the license enablers if the NetWorker configuration files still exist. By default, the configuration files are located in the SYS:NSR\RES directory.

2. Reinstall any NetWorker patches that were installed prior to the disaster.

Task 2: Configure NetWorker Device and Client Resource

To configure the Device and Client resource:

1. Configure the Device resource:

Note:

If you are recovering data by using an autochanger, press [F4] to verify that the autochanger correctly appears in the Autochanger window.

2. Configure the Client resource associated with the server.

Set the browse and retention policies to a decade. The browse policy is one quarter (three months) by default. This enables you to recover all of the server’s records.

Important:

If you do not reset the browse and retention policies, all of the server’s records will be recovered. However, any records that are more than one quarter (three months) old will be discarded, because the browse policy is one quarter by default.

108

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

Task 3: Locate the Server’s Bootstrap Save Set ID

This section describes how to locate the save set ID for the NetWorker server’s latest bootstrap. The NetWorker server’s bootstrap contains the server’s online file index, media index, and resource files.

If you do not have this information, use the following steps to locate the save set ID of the most recent bootstrap:

1. Insert the most recent media or clone volumes used for scheduled backups into the appropriate device.

2. At the NetWare system console, switch to the directory where the

NetWorker NLMs are located. By default, these files are located in the

SYS:NSR\BIN directory.

3. To locate the most recent bootstrap on the media, enter the scanner -B command.

Task 4: Recover the NetWorker Server Bootstrap

This section describes how to recover the NetWorker server’s bootstrap. By default, these files reside in the SYS:NSR and SYS:NSR\RES directories, and can be recovered by using the NetWorker Utilities program.

To recover the NetWorker bootstrap:

1. Load and inventory the devices. This ensures that the NetWorker software can recognize which slots contain which volumes.

If you load a clone volume into a device, the NetWorker software will use the clone volume for the remainder of the recovery process. You might have to update the volume location to indicate that the original volume is not available.

2. Start the NetWorker Utilities program by entering the following command at the NetWare system console:

load NETUTIL

Important:

Disable scheduled backups and do not use the NetWorker software to perform backups or recoveries while running the NetWorker Utilities program. If backup and recovery processes are enabled, the state of the

NetWorker indexes will become confused. From the NetWorker Utilities dialog box, select Recover from a Disaster.

6

NetWorker V7.3

Disaster Recovery Guide

109

6

Chapter 6: NetWare Disaster Recovery

3. Enter the full name of the backup administrator (for example,

“.cn=Admin.O= ‘top level container’”) into the User text box and the appropriate NetWare password into the Password text box and press

[Enter].

The Device Selection window appears.

4. From the Device Selection window, select the device you intend to use and press [Enter].

5. Insert the most recent media or clone volumes used for scheduled backups into the appropriate device.

• If you are using a single tape device, manually insert the volume.

• If you are using an autochanger, use the autochanger controls to manually select the slot containing the most recent backup volume.

To determine the required backup volume, refer to the printed bootstrap records.

6. When the following message appears, press [Enter].

Put the volume for disaster recovery in device xxxx and press Enter.

7. Press [Enter] to exit the NetWorker Utilities. The server’s processor will halt with multiple ABENDs.

8. Restart the NetWorker software.

9. From the NetWorker Utilities program, select F3>Operation>Recover.

10. Browse and mark the following files for recovery:

• SYS:NSR\NSR.RES

• SYS:NSR\NSRJB.RES

11. Press [F2] to start the recovery. When messages indicating name conflicts appear, select Rename the Recover File for both files.

12. When recovery is complete, exit the NetWorker Utilities program.

110

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

Task 5: Rename the NetWorker Server Resource Files

To rename the resource files:

1. Unload all of the associated NLM files by using nwdown.

2. Delete the existing files (NSR.RES and NSRJB.RES).

3. Rename the recovered files, which were renamed with a tilde (~) upon recovery, to NSR.RES and NRJB.RES.

4. Restart the NetWorker software. This process recovers the NetWorker software to its last backup configuration (including passwords, administrator privileges, backup groups, and schedules).

5. Verify that the NetWorker indexes and configuration files are recovered. If they are not recovered, select Recover from a Disaster from the NetWorker

Utilities program, or use an older volume.

Task 6: Recover the NetWorker Server Data

To recover the NetWorker server data:

1. Insert the most recent backup or clone volumes into the device.

2. From the Mount dialog box, select the required volume.

Notice that the volume is now marked (R); this means the volume is write protected.

3. From the [F3] Operation menu, select Recover to display the Client List window.

4. Select the client with the NetWare server name and press [Enter].

Another login window appears.

5. Enter the appropriate information for each attribute and press [Enter]:

• In the User attribute, enter the full name of the backup administrator

(for example, “.cn=Admin.O= ‘top level container’”).

• In the Password attribute, enter the NetWare password.

The Browser window appears.

6. Select Schema and mark it for recover so that all extensions to the NDS schema are recovered.

7. From the [F2] File menu, select Start Recover, and press [Enter].

8. If you have NDS data to recover, continue with

"Task 8: Recover NDS

Data" on page 112 .

NetWorker V7.3

Disaster Recovery Guide

111

6

6

Chapter 6: NetWare Disaster Recovery

If you do not have any NDS data to recover, continue with

"Task 9:

Complete the Server Recovery" on page 113 .

Task 7: Create a Record of Backups Done after Bootstrap Created

After a disaster recovery, the media database will not contain information about any backups done after the bootstrap was created. The media database retains information about the last known file/record marks on the tape, and if backups were done after the bootstrap was created, the file/record information will not be current. For any tapes that contain backups that were completed after the bootstrap was created, use scanner with, at a minimum, the -m option to scan the save sets and/or tapes into the media database. This will ensure that the media database has a record of all completed savesets and that valid backups will not be overwritten by future backups.

Additionally, write protect all appendable volumes until you can complete the scanner activity to prevent scheduled backups from using these volumes. See

"Task 8: Create a Record of Backups Done after Bootstrap Created" on page 37

for instructions on write protecting and scanning.

112

Task 8: Recover NDS Data

Before performing an NDS backup or recover, refer to the Novell Application

Notes section “Backing Up and Restoring Novell Directory Services in

NetWare 4.11,” dated October 1996. This document explains concepts and procedures that you need to understand before working with NDS backups and recovers.

On networks with multiple servers, NDS automatically creates replicas

(copies) of the NDS database or portions of it (partitions) and stores them on other servers. This process provides a readily available backup if NDS or a partition is damaged. Do not circumvent this replication process.

Whenever possible, use an active replica to recover what was lost from the

NDS tree. If this is not feasible, you must recover from a Storage Management

Services (SMS) backup in the following way:

1. Recover the NDS information.

2. Recover the file system data and trustee rights.

NDS backup and recover is based on object names. The objects must exist in the tree before you can recover the file system data and trustee assignments for those objects. NDS should be functional (time and partitions synchronizing normally) before you proceed with a recovery.

To recover NDS data to this server by using the NetWorker software:

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

• Create a backup of the NDS root rather than a backup of only NDS leaf objects.

• Use the Recover or Directed Recover option from the Operation menu.

To recover NDS data

1. From the [F3] Operation menu, select Save Set Recover, and press [Enter].

Note:

The Save Set Recover option in the NetWare version of the

NetWorker Administrator program can not be used to restore NDS data if the original backup contained only leaf objects.

2. Select this server’s name as both the source and destination client.

3. Enter the full name of the backup administrator, and the password.

4. Select the {NDS} save set, and press [Enter] to display versions.

5. Select the version with the most recent full backup.

If there are any later save sets with level 1 or incremental backups, recover them from the oldest to the most recent.

The NDS is now recovered to the state of the most recent backup.

Task 9: Complete the Server Recovery

To complete the recovery of the server:

1. Use NetWorker software to recover the remaining data, including the client file indexes. In the Browser window, mark everything for recovery, and then unmark all still-operational volumes, LEGATOMM, the server’s

LEGATODB, NSR.RES, NSRJB.RES, Schema, and NDS.

Important:

Recover each client file index by selecting the client folder from the indexes directory. Each client has a legatodb file that is located in

SYS:NSR\INDEX\CLIENT-NAME.

If you run out of memory while recovering multiple files, repeat this step and recover only one volume at a time.

2. From the [F2] File menu in the Browser window, display the NetWorker

Recover Options dialog box. Select Don’t Overwrite Data.

6

NetWorker V7.3

Disaster Recovery Guide

113

6

Chapter 6: NetWare Disaster Recovery

3. Verify the recovered data. From a workstation, use the ndir command or

NetWare Administrator (NWAdmin) to check the data, trustee assignments, file ownership, and other related information.

4. Select Cross-Check Index from the Indexes dialog box of the NetWorker

Administrator program to compare the index records to the records in the media index.

5. Perform a check to make sure the server is fully recovered:

• Verify that the server and its associated clients are included in a scheduled backup.

• Run dsrepair to verify the integrity of any NDS databases.

• Use the NetWorker Administrator program or the ndir workstation utility to verify the data, trustee assignments, and file ownership.

Recovering a Replicated NDS Partition

If the NetWare server did not contain an NDS partition (replicated or not), do not complete this section. Instead, complete only

"Recovering the NetWare

Operating System" on page 103 and

"Recovering a NetWare NetWorker

Server" on page 107 .

This section contains the following information on recovering a replicated

NDS partition over the network:

"How to Recover a NetWare 4.10 SYS Volume" on page 115

"How to Recover a NetWare 4.11, NetWare 5.0, or IntraNetWare Server

SYS Volume" on page 118

Important:

For NetWare 4.11/IntraNetWare servers, do not delete the server or volume objects for the failed volume from the NDS tree; you do not want to eliminate any references other objects might have to the volume. If you must delete objects on a NetWare 4.11/IntraNetWare server, use the NetWare 4.10 procedure for recovering from a disaster.

114

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

How to Recover a NetWare 4.10 SYS Volume

To recover a SYS volume on a NetWare 4.10 server, complete the following tasks:

"Task 1: Recover the NetWare 4.10 SYS Volume" on page 115

"Task 2: Recover the SMS Remote File System" on page 116

"Task 3: Complete the Recovery of the NetWare 4.10 SYS Volume" on page 117

Task 1: Recover the NetWare 4.10 SYS Volume

To recover a SYS volume on a NetWare 4.10 server:

1. Use the NetWorker Administrator program or netadmin to delete the volume objects associated with the failed server.

2. Use NDS Manager or partmgr to delete the server object for the failed server. You cannot use netadmin to delete a server object.

Partition Manager displays a warning message; select Yes to confirm the deletion.

3. Use NDS Manager or dsrepair to check the replica synchronization.

If you see error messages, wait a few minutes and try again.

4. From the NetWorker server, perform a directed recover to recover the failed server’s Server Specific Info (SSI) files from a tape backup to a functioning NetWorker for NetWare client.

The server-specific information files (SERVDATA.NDS, VOLSINFO.TXT,

STARTUP.NCF, and AUTOEXEC.NCF) are recovered to a subdirectory under SYS:\SYSTEM on the server you selected. This subdirectory is given a DOS 8.3 name derived from the source server name.

5. If the failed server held a master replica, use NDS Manager or dsrepair to designate a new master replica on a different server in the replica ring.

6. Use NDS Manager or dsrepair to perform an unattended full repair to check replica synchronization. If necessary, use NDS Manager or dsrepair on the servers containing master replicas to remove the failed server from the replica ring.

7. Shut down the failed server and replace any damaged server hardware. If you replace a hard drive, be sure that it is the same size or larger.

8. Format the DOS partitions and reinstall DOS.

NetWorker V7.3

Disaster Recovery Guide

115

6

Chapter 6: NetWare Disaster Recovery

6

116

Important:

Use the DOS Time command to ensure that the computer is set to the correct time to avoid time synchronization errors.

9. Reinstall NetWare 4.10 and NDS on the repaired or replaced server. Run

install

or nwconfig, select Custom Install, and follow the directions on the screen. Use the STARTUP.NCF and AUTOEXEC.NCF files recovered with

SSI to answer the questions displayed on the screen.

– Enter the same server name and internal IPX number that the server had prior to the disaster.

– When prompted, insert the NetWare License diskette for the server into the disk drive.

– When prompted for the name of the NDS tree, select the name of the tree that the server resided in before the disaster.

– Select the time zone and configure the time.

– Log in and specify the context for the server and its objects. Use the same context that was used before the disaster.

– Edit the STARTUP.NCF and AUTOEXEC.NCF files to match the versions recovered with SSI.

When the installation is complete, the server will contain all the files necessary to perform an SMS remote file-system recover. For more information, see

"Task 2: Recover the SMS Remote File System" on page 116 .

Task 2: Recover the SMS Remote File System

To recover the SMS remote file system to a SYS volume on a NetWare 4.10 server:

1. Load the required namespace modules for each recovered volume. Use the

VOLSINFO.TXT file to determine which namespaces need to be loaded

(MAC.NAM, LONG.NAM, etc.).

2. Load the file-system TSA specific to your version of the NetWare operating system, by entering one of the following commands:

LOAD TSA410

LOAD TSA312

LOAD TSA500

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

3. Recover the file system for each volume affected by the failure. Do not recover the Schema and [Root]; they are recovered from a replica. You also do not need to recover Server Specific Info again. When prompted, suppress further prompting and overwrite files.

4. When you are prompted, log in by using the full name of the backup user.

5. If the failed server had non-SYS volumes that were not affected by the failure, from the [F2] File menu in the Browser window, select Recover to display the NetWorker Recover Options dialog box. Select the Don’t

Overwrite Data; restore trustees, etc.

command, and then recover the volumes that were not affected by the failure.

6. Shut down and restart the repaired or replaced server.

Task 3: Complete the Recovery of the NetWare 4.10 SYS Volume

To complete the recovery of a NetWare 4.10 SYS volume:

1. If necessary, use NDS Manager or dsrepair to reestablish replicas on the repaired or replaced server.

2. Enter the following commands at the command prompt:

LOAD TSA410

LOAD TSANDS

LOAD TSA500

3. From the Recover Browser window, recover the server object, volume objects, and any objects that formerly referenced the recovered volume or server objects. Expand the [Root] resource, mark the required objects, and then select Recover. When prompted, suppress further prompting and overwrite files.

4. Use NDS Manager or the Schedule immediate synchronization function of

dsrepair

to synchronize the replica on all servers.

5. Verify the recovered data. From a workstation, use either the nwadmin server utility or the ndir workstation utility to check the data, trustee assignments, file ownership, and other related information.

The SYS volume should now be recovered.

6

NetWorker V7.3

Disaster Recovery Guide

117

6

Chapter 6: NetWare Disaster Recovery

How to Recover a NetWare 4.11, NetWare 5.0, or IntraNetWare Server

SYS Volume

To recover a SYS volume on a NetWare 4.11, NetWare 5.0, or IntraNetWare server, complete the following tasks:

"Task 1: Recover the NetWare SYS Volume" on page 118

"Task 2: Reinstall the Operating Systems" on page 119

"Task 3: Complete the NetWare SYS Volume Recovery" on page 121

Important:

For NetWare 4.11/IntraNetWare servers, do not delete the server or volume objects for the failed volume from the NDS tree; you do not want to eliminate any references other objects might have to the volume. If you must delete objects on a NetWare 4.11/IntraNetWare server, use the NetWare 4.10 procedure for recovering from a disaster.

Task 1: Recover the NetWare SYS Volume

To recover a NetWare 4.11 or IntraNetWare SYS volume:

1. From the NetWorker server, perform a directed recover to recover the failed server’s Server Specific Info (SSI) files from a tape backup to a functioning NetWorker for NetWare client.

The server-specific information files (SERVDATA.NDS, DSMISC.LOG,

VOLSINFO.TXT, STARTUP.NCF, and AUTOEXEC.NCF) are recovered to a subdirectory under SYS:\SYSTEM on the client you have selected. This subdirectory is given a DOS 8.3 name derived from the source server name.

Important:

For NetWare 4.11/IntraNetWare servers, do not delete the server or volume objects for the failed server from the NDS tree. You do not want to eliminate any references other objects might have to the server. If objects were deleted from the NDS tree, use the NetWare 4.10 procedure for recovering from a disaster.

118

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

2. If the failed server held a master replica, use NDS Manager or dsrepair to designate a new master replica on a different server in the replica ring. To determine which replicas were stored on the failed server, refer to

DSMISC.LOG.

3. If the failed server also contained any nonmaster replicas, use NDS

Manager or dsrepair on the servers containing master replicas to remove the failed server from the replica ring.

A NetWare warning message appears. Continue with the recovery procedure. For more information, refer to “Backing Up and Restoring

Novell Directory Services in NetWare 4.11” in Novell Application Notes,

October 1996.

4. Use dsrepair to perform an unattended full repair to ensure the ring is functioning properly.

Refer to DSMISC.LOG to determine which replicas were stored on the failed server. If DSMISC.LOG shows that no other server has exactly the same replicas as the failed server, run dsrepair on any servers containing replicas of partitions on the failed server.

5. Shut down the failed server and replace any damaged hardware. If you need to replace a hard drive, install a drive that is the same size or larger.

Task 2: Reinstall the Operating Systems

To reinstall the operating systems:

1. Format the DOS partitions and reinstall DOS.

Important:

Use the DOS Time command to ensure that the computer is set to the correct time to avoid time synchronization errors.

2. Reinstall NetWare 4.11, NetWare 5.0, or IntraNetWare and NDS on the repaired or replaced server. Run install or nwconfig, select Custom

Install, and follow the directions on the screen.

a. When prompted, enter the same server name and internal IPX number that the server had prior to the failure. Use the

STARTUP.NCF and AUTOEXEC.NCF files included with the server-specific information for needed information.

6

NetWorker V7.3

Disaster Recovery Guide

119

6

Chapter 6: NetWare Disaster Recovery b. After the preliminary files are copied, the Choose a Directory Tree dialog box appears. Press [F5] to recover NDS (option listed at the bottom right of the screen).

c. A new window displays two options: A: (the default) or Press [F3] to specify a different path. If the Server Specific Info files are contained on diskette, insert the diskette into drive A and press [Enter]. Otherwise, press [F3] and enter the path to the Server Specific Info files recovered in step 1. d. A Remote Server Authentication login dialog box is displayed. Log in. When prompted, enter the Directory tree name.

e. Press [Enter], and both the files and NDS are copied to the new server. DSMISC.LOG, VOLSINFO.TXT, and AUTOEXEC.NCF are copied to the SYS:SYSTEM directory. STARTUP.NCF is copied to the

C:\NWSERVER directory.

f. The NDS recovery uses the information from SERVDATA.NDS

(TSANDS.NLM is not needed). NDS is now fully functional on the server, but the partitions and replicas must still be reestablished.

g. When prompted, insert the NetWare License diskette for the server into the diskette drive.

h. Edit the STARTUP.NCF and AUTOEXEC.NCF files.

i. If either the STARTUP.NCF or the AUTOEXEC.NCF files have changed because they were backed up with the server-specific information, both the original and the new files are displayed for you to compare and make edits as necessary. If the current files are the same as the original files, only the current files are displayed.

The server now contains all the files necessary to perform an SMS remote file system recover.

3. To finish the installation, do one of the following:

• Press [Enter] to exit the utility. NetWare will not copy the remaining system and public files but will exit the utility. These files should be recovered from a backup.

• Press [F3] to continue installation and wait while the utility copies the remaining system and public files. Then exit.

120

NetWorker V7.3

Disaster Recovery Guide

Chapter 6: NetWare Disaster Recovery

Task 3: Complete the NetWare SYS Volume Recovery

This section describes how to complete the recovery of the 4.11 SYS volume.

To complete the recovery of the 4.11 SYS volume:

1. Load the required namespace modules for each recovered volume. Use the

VOLSINFO.TXT file to determine which namespaces need to be loaded

(MAC.NAM, OS2.NAM, and so on).

2. Load the file-system TSA specific to your version of NetWare by entering one of the following commands on the repaired or replaced server:

LOAD TSA410

LOAD TSA500

3. Recover the file system for each volume affected by the failure. Do not recover the Schema and [Root]; they will be recovered from a replica. You also do not need to recover Server Specific Info again. When prompted, suppress further prompting and overwrite files.

If the failed server had non-SYS volumes that were not affected by the failure, no further action is needed because the SERVDATA.NDS file preserves the trustee assignments on these other volumes.

4. Shut down and restart the server.

5. Use NDS Manager or dsrepair to re-establish replicas on the failed server.

Use DSMISC.LOG to view a copy of the replica list that resided on the server at the time of backup.

6. Verify the recovered data. From a workstation, use the Novell

NWAdmin32

or the rights /T /S and ndir commands to check the data, trustee assignments, file ownership, and other related information.

6

NetWorker V7.3

Disaster Recovery Guide

121

Chapter 6: NetWare Disaster Recovery

6

122

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery

(Windows NT 4.0)

This chapter describes how to recover one or more computers in a Windows

NT 4.0 cluster.

This chapter includes the following sections:

"Recovering One Node in a Cluster" on page 123

"Recovering a Cluster Shared Disk" on page 125

"Replacing the Quorum Disk and Recovering Its Data" on page 128

"Recovering the MSCS Cluster Configuration" on page 129

"Performing a Cluster-Wide Recovery" on page 131

Important:

To ensure proper termination, never power off any node in the cluster unless it is connected with a Y cable.

Recovering One Node in a Cluster

Note:

In the example used for the following tasks, Node_A is still working and

Node_B has failed.

To recover a failed node within a cluster, perform the following tasks:

"Task 1: Evict Node_B from the Cluster" on page 124

"Task 2: Reinstall Windows NT 4.0 and MSCS on Node_B" on page 124

"Task 3: Reinstall the NetWorker Software" on page 125

"Task 4: Recover the Data on Node_B" on page 125

NetWorker V7.3

Disaster Recovery Guide

123

7

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

Task 1: Evict Node_B from the Cluster

To evict Node_B from the cluster:

1. Ensure that the cluster resources failed over to Node_A, the working node.

This includes the quorum disk and any shared disks.

2. If any of the data that failed over from Node_B to Node_A is corrupt, use the NetWorker software to recover the data.

Note:

You may need to shut down some cluster resources that depend on this data.

3. From the Cluster Administrator on Node_A, evict Node_B from the cluster.

4. Shut down Node_B if it is still running and disconnect it from the cluster.

5. Fix or replace the damaged hardware on Node_B, if necessary.

6. Reconnect Node_B to the cluster.

Task 2: Reinstall Windows NT 4.0 and MSCS on Node_B

To reinstall Windows NT and MSCS on Node_B:

1. Reinstall the Windows NT 4.0 Server Enterprise Edition.

2. Apply Windows NT 4.0 service pack 3.

3. Reinstall the MSCS software.

4. If the node had a local tape drive, reinstall the device driver.

5. Reinstall the latest Windows NT 4.0 service pack.

6. Join Node_B to the cluster with Node_A.

7. Assign the same drive letters that were previously assigned to the disk volumes.

Important:

The assigned local drive letters must not conflict with the drive letters assigned to any of the shared disks.

124

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

Task 3: Reinstall the NetWorker Software

To reinstall the NetWorker software, on Node_B:

1. Log in to the same domain as Node_A.

2. Install the NetWorker software. For instructions, refer to the EMC

NetWorker Installation Guide, Microsoft Windows Version.

Task 4: Recover the Data on Node_B

To recover the data on Node_B:

1. Recover all the hives in the Registry or system state.

2. Use NetWorker to recover any data that is local only to Node_B.

3. Restart Node_B.

4. Verify that the cluster is working properly.

Recovering a Cluster Shared Disk

This section describes how to recover data from a cluster shared disk that is not a quorum disk.

To recover a cluster shared disk, perform the following tasks:

"Task 1: Take the Dependent Resources Offline" on page 126

"Task 2: Replace the Hard Drive" on page 126

"Task 3: Recover the Data to the Shared Disk" on page 127

"Task 4: Configure the Cluster to Include the Shared Disk" on page 127

NetWorker V7.3

Disaster Recovery Guide

125

7

7

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

Task 1: Take the Dependent Resources Offline

Note:

The following procedure outlines the required steps for taking the dependent resources offline. For detailed instructions, refer to the Microsoft

Cluster Server documentation.

To take offline all of the resources that are dependent on the shared drive that you are replacing:

1. In the Cluster Administrator program, create a new resource group.

2. If the current shared disk’s resource group contains the shared disk and its dependent resources, move the cluster resources that represent the affected shared disk into the resource group created in the previous step.

3. Take offline the resource group that contains the shared disk.

Note:

If the shared disk’s resource group only contains resources that depend on it, there is no need to create a new resource group.

Task 2: Replace the Hard Drive

For more information, refer to the Microsoft Cluster Server documentation.

To replace the hard drive:

1. Using the Control Panel Services tool, set the Startup Type of the Cluster

Service to Manual.

2. In the Cluster Administrator program, open the Property window for each resource in the resource group. Select the Do Not Restart property from the

Advanced window.

3. Remove the shared disk from the dependency list of any resources that depend on the shared disk.

126

Important:

If you fail to perform this step, you may not be able to delete the disk resource in the following step.

4. Delete the cluster resource that represents the shared disk.

5. If required, shut down the nodes.

Note:

Refer to the MSCS Cluster Hardware Manual to determine whether you must shut down the nodes before you replace the shared disk.

6. Remove the shared disk and replace it with a new shared disk.

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

7. If you have shut down the nodes, reboot them.

8. Create the drive volumes and format the new disk as the old one was.

Task 3: Recover the Data to the Shared Disk

To recover the data that belongs to the shared disk:

1. Ensure that the NetWorker software is properly installed. For more information, refer to the EMC NetWorker Installation Guide, Microsoft

Windows Version.

2. Using the NetWorker User program, recover all the data that belongs to the shared disk.

Note:

If the resource group is a virtual server, you may need to perform a directed recovery. For more information, refer to the NetWorker Administrator’s

Guide.

Task 4: Configure the Cluster to Include the Shared Disk

For more information, refer to the Microsoft Cluster Server documentation.

To configure the cluster to include the shared disk:

1. Using the Control Panel Services tool, start the Cluster Service if it was previously stopped.

2. In the Cluster Administrator program, create a new cluster resource to represent the new disk.

3. Add the cluster resource to the resource group you created in "Task 1: Take the Dependent Resources Offline" on page 126 .

If you did not create a new resource group in Task 1, add the resource to the group where it was before the replacement.

4. Add the shared disk to the dependency list of all the cluster resources that depend on this resource.

5. Start the resource group that contains the shared disk.

6. If you created a new resource group: a. Move the disk resource and all dependant resources to their original resource group.

b. Delete the now-empty resource group.

7

NetWorker V7.3

Disaster Recovery Guide

127

7

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

7. Ensure that the Startup Type of the Cluster Service is set to Automatic.

8. Ensure that the resource group property is set to Do Not Restart. In most situations, it is unselected.

Replacing the Quorum Disk and Recovering Its Data

This section provides information about the Quorum resource and explains how to recover a quorum disk’s data.

About the Quorum Resource

If there is a communication failure between the nodes, the MSCS software uses the Quorum resource in an arbitration process to determine the members of the cluster nodes. The Quorum resource also records changes made to the cluster database when one of the nodes in the cluster is down.

Refer to the Microsoft Cluster Server documentation for detailed information on how to:

• Locate the quorum disk

• Locate the quorum data directory

• Change the quorum disk designation

Recovering the cluster configuration data on the quorum disk is important during a disaster recovery. Recovery of the Quorum resource information is not automatic.

• If the quorum disk is owned by a virtual server, configure the virtual server as a NetWorker client. For the client’s Save Set attribute, specify one of the following:

– All

– The disk volume or directory that contains the quorum data

• If the quorum disk is not owned by a virtual server, configure all the nodes as NetWorker clients. For the client Save Set attributes, specify one of the following:

– All (this is preferred)

– The disk volume or directory that contains the quorum data

Note:

The quorum disk is accessible from only one node at a time. Therefore, the backup will fail over to the node that does not have access to the quorum disk. This is why specifying save set All is preferred over specifying the disk volume or directory that contains the quorum data.

128

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

How to Replace the Quorum Disk and Recover Its Data

To recover the quorum disk’s data:

1. Take offline any cluster resources or applications that depend on the quorum disk’s data.

Important:

Do not take the Cluster Service offline.

2. Change the quorum disk’s drive letter and replace the quorum disk with a new disk.

Note:

For detailed instructions, refer to the appropriate Microsoft documentation. In particular, refer to Microsoft Knowledge Base article

Q280353, How to Change Quorum Disk Designation.

3. If necessary, install the NetWorker software on the node that has access to the quorum disk.

4. Recover any data except the quorum data that resided on the quorum disk prior the disaster.

Important:

Do not recover the quorum data.

Recovering the MSCS Cluster Configuration

The following procedure is based on these assumptions:

• You have not changed the disk configuration from the time you performed the last backup on the cluster database.

• You have not replaced any hardware, including the shared disk, from the time you performed the last backup on the cluster database.

• Node_A has the latest backup of the system state.

• Node_A or a virtual server has the latest backup of the quorum data.

7

NetWorker V7.3

Disaster Recovery Guide

129

7

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

To recover the MSCS cluster configuration, perform the following tasks:

"Task 1: Recover the System State and Quorum Data" on page 130

"Task 2: Replace the Quorum Data with the Recovered Quorum Data" on page 130

"Task 3: Evict Node_B from the Cluster" on page 131

"Task 4: Reinstall the Cluster Software and Patches on Node_B" on page 131

Task 1: Recover the System State and Quorum Data

To recover the system state and quorum data:

1. Select a node in the cluster that has the good (and possibly the latest) backup of the system state and the quorum data. (This procedure uses

Node_A, for example.)

2. Verify that the Cluster Service is running on Node_A.

3. From Node_A, recover the following:

• System state

• Quorum data (If the quorum disk is owned by a virtual server, use the directed recover procedure. For more information, refer to the

NetWorker Administrator’s Guide.)

Make sure that you recover the quorum data to a different directory on the quorum drive.

For example, if the quorum data previously resided in G:\MSCS, recover the data to G:\RECOVER\MSCS.

4. Shut down Node_A.

Important:

Do not reboot Node_A.

Task 2: Replace the Quorum Data with the Recovered Quorum Data

To replace the quorum data to Node_B:

1. On Node_B, uninstall the cluster software.

2. Reboot Node_B.

130

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

3. Copy the existing quorum data from its original location to another directory. For example, copy:

G:\MSCS

to

G:\TEMP\MSCS

4. Replace the existing quorum data with the recovered quorum data. For example, copy:

G:\RECOVER\MSCS

to

G:\MSCS

5. Shut down Node_B.

Task 3: Evict Node_B from the Cluster

To evict Node_B from the cluster:

1. Reboot Node_A.

2. Open the Cluster Administrator and evict Node_B from the cluster.

Task 4: Reinstall the Cluster Software and Patches on Node_B

To reinstall the cluster software and patches on Node_B:

1. Reboot Node_B.

2. Reinstall the cluster software and join it to the cluster in which Node_A is a member node.

3. On Node_B, apply any needed cluster patches and Windows service packs.

4. Verify that the cluster configuration has been successfully recovered.

Performing a Cluster-Wide Recovery

Use the following procedures to perform a complete cluster recovery, including reinstallation of the operating system on all nodes. These procedures describe how to perform a recovery, either by using existing hardware or new hardware.

7

NetWorker V7.3

Disaster Recovery Guide

131

7

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

In the following example:

• Choose the node that has the most up-to-date backup of the cluster database (part of the Registry) to serve as Node_A.

• Defective nodes and shared drives have been replaced.

To recover the cluster to new hardware or existing hardware, perform the following tasks:

"Task 1: Reinstall and Configure the Software on Node_A" on page 132

"Task 2: Reinstall and Configure the Software on Node_B" on page 133

"Task 3: Reinstall and Configure NetWorker" on page 133

"Task 4: Perform a Disaster Recovery" on page 134

"Task 5: Check and Reassign the Drive Assignments" on page 134

"Task 6: Recover the Data" on page 135

"Task 7: Check the Disk Signatures" on page 136

"Task 8: Join Node_B to the Cluster" on page 137

Task 1: Reinstall and Configure the Software on Node_A

To reinstall and configure the software on Node_A:

1. Make sure the node has been properly:

• Shut down.

• Connected through the shared disks.

• Terminated.

2. Reinstall Windows NT 4.0 Server Enterprise Edition.

a. Using the same cluster log on account, add the server to the domain.

b. Create the partitions so that the new disk configuration of Node_A matches the original configuration.

c. Use the same IP address and server name (fully-qualified DNS domain name, if used) that were used originally.

d. Using the original settings whenever possible, configure the network adapters and the IP address.

e. Log on to the domain.

3. Using the Windows NT Disk Administrator, configure the drives to match their previous configuration.

4. Reinstall Windows NT 4.0 Server Enterprise Edition server-specific service packs (Service Pack 3).

132

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

5. Reinstall the MSCS software.

6. Reinstall any required service pack software. Make sure that the version is later than Service Pack 3.

7. If Node_A is a NetWorker server, configure the tape devices.

8. Using the cluster account on the domain, log on to Node_A.

Task 2: Reinstall and Configure the Software on Node_B

To reinstall and configure the software on Node_B:

1. Make sure the node has been properly:

• Shut down.

• Connected through the shared disks.

• Terminated.

2. Reinstall Windows NT 4.0 Server Enterprise Edition.

a. Using the same cluster log on account, add the server to the domain.

b. Create the partitions so the new disk configuration matches the original configuration.

c. Use the same server name (fully-qualified DNS if used) and IP address that were used originally.

d. Using the original settings whenever possible, configure the network adapters and the IP address.

e. Log in to the domain.

3. Using the Windows NT Disk Administrator, configure the drives to match their previous configuration.

4. Reinstall Windows NT 4.0 Server Enterprise Edition server-specific service packs (Service Pack 3).

Task 3: Reinstall and Configure NetWorker

Reinstall and configure the NetWorker software and any required patches on

Node_A and Node_B. For more information, refer to the EMC NetWorker

Installation Guide, Microsoft Windows Version.

7

NetWorker V7.3

Disaster Recovery Guide

133

7

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

Task 4: Perform a Disaster Recovery

To perform a disaster recovery:

1. Shut down Node_B.

2. Perform a NetWorker disaster recovery of Node_A. For more information, see

"Chapter 4: Windows Disaster Recovery" on page 49

.

3. Recover the following directories:

• Windows NT Registry or system state

• Windows NT system directory

Select Overwrite and Suppress Messages.

4. Recover the quorum and the shared drives data. Follow the directed recovery procedure. For detailed instructions, refer to the NetWorker

Administrator’s Guide.

Note:

Make sure you recover the \MSCS directory in the quorum to a different location, for example \MSCStemp.

5. Shut down Node_A. Do not reboot Node_A.

a. The system will prompt you to restart. Select Yes.

b. Shut down the node when the system is in its reboot cycle.

6. Restart Node_B and log in with Administrator privilege.

7. On the quorum drive: a. Rename the existing \MSCS directory to \MSCSold.

b. Move the recovered \MSCS directory that was recovered to

\MSCStemp to the \MSCS directory.

8. Shut down Node_B.

Task 5: Check and Reassign the Drive Assignments

If you are replacing the cluster shared disks with new hardware, follow these steps to check and reassign the drive assignments:

1. Reboot Node_A. The drive assignments may have changed. Use the Disk

Administrator to reassign the drives to their former drive letters.

2. Using the Control Panel Services tool, set the Cluster Service to Manual startup.

3 . Reboot Node_A.

134

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

4. Replace the cluster shared disks.

5. Check the drive assignments through Windows Explorer. The “phantom” drives you see are the original shared disks. These drives cannot be deleted or disconnected from the Windows Explorer.

6. Remove the Registry entries for the phantom disks: a. Use regedit to display the following Registry entry:

HKEY_LOCATION_MACHINE\SYSTEM\CurrentControlSet\Services

\ Clusdisk\Parameters\Signatures

The numbers listed under Signatures are the drive signatures for the former shared disks. Delete these numbers.

b. Reboot the node.

c. Using Windows Explorer, check the status of the disks. There should no longer be any phantom disks and the new shared disks should have the proper drive letter assignments.

The new shared disk signatures are located in the following Registry key:

HKEY_LOCATION_MACHINE\SYSTEM\CurrentControlSet\Services

\Clusdisk\Parameters\AvailableDisks

Task 6: Recover the Data

To recover the data:

1. Use the NetWorker User program to recover the remaining data from tape.

Do not recover the following:

• System state

• RepairDisk

• Windows NT system directory

For more information, see "Chapter 4: Windows Disaster Recovery" on page 49 .

2. Reboot Node_A.

3. Log in to a domain with Administrator privileges.

7

NetWorker V7.3

Disaster Recovery Guide

135

7

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

Task 7: Check the Disk Signatures

If you are replacing the shared disks, perform the following procedure to check the disk signatures. If you are not replacing the shared disks, skip this task and proceed to

"Task 8: Join Node_B to the Cluster" on page 137 .

To check the disk signatures:

1. Replace the cluster shared disks with the new hardware: a. Using the Control Panel Services tool, start the Cluster Service with the -fixquorum startup option. This option allows you to fix or add a new quorum drive.

b. Start the Cluster Administrator program and attach Node_A to the cluster you are recovering.

c. Create the new disk resource for the quorum drive and place it in the

Quorum Group.

Note:

For detailed information, refer to Microsoft Knowledge Base article Q280353: How to Change Quorum Disk Designation.

d. From the Cluster Administrator program:

– Rename all of the old shared disks.

– Create new resources for the physical disks to match the old shared disks.

– Check for dependencies for the physical disk and make the necessary modifications. e. Create a new group and move all of the old shared disks to this group. (This is done for cleanup purposes.)

Important:

You cannot delete the old shared disks until both nodes are running and both nodes are members of the cluster.

2. Using the Control Panel Services tool, set the MSCS service to Automatic startup.

3. Remove any Startup Parameters that may have been previously set. For example, remove -fixquorum.

136

NetWorker V7.3

Disaster Recovery Guide

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

Task 8: Join Node_B to the Cluster

To join Node_B to the cluster:

1. On Node_A, use the Cluster Administrator to evict Node_B from the cluster.

2. Restart Node_B.

3. On Node_B, reinstall the MSCS software and rejoin the cluster.

4. Reinstall the latest service pack software on Node_B. Make sure the version is greater than Service Pack 3.

5. If NetWorker Server is reinstalled in Node_A as a failover server: a. Reinstall NetWorker server software.

b. Register the resource extension. This manages the NetWorker server resource type in Node_B. For more information, refer to the EMC

NetWorker Installation Guide, Microsoft Windows Version. c. On Node_B, you can choose to:

– Configure the tape devices.

– Reinstall the NetWorker client software.

6. Reinstall any NetWorker specific patches.

7. Recover all the data that belongs to Node_B. This includes the system configuration data. For example:

• System State

• RepairDisk

• Windows NT 4.0 %SystemRoot% directory

• C:

8. Reboot Node_B.

9. Verify that the system has been recovered to its original state before the disaster.

7

NetWorker V7.3

Disaster Recovery Guide

137

Chapter 7: Microsoft Cluster Disaster Recovery (Windows NT 4.0)

7

138

NetWorker V7.3

Disaster Recovery Guide

Chapter 8: Microsoft Cluster Disaster Recovery

(Windows 2000 and 2003)

This chapter explains how to recover Microsoft Windows 2000 and Windows

2003 Microsoft Cluster Server (MSCS) host systems. Familiarity with MSCS concepts and operations is assumed. For complete information about MSCS, refer to the Microsoft documentation.

This chapter includes the following sections:

"Prerequisites" on page 139

"Recovering a Failed Quorum Disk" on page 140

"Recovering One Cluster Node" on page 141

"Recovering Multiple Cluster Nodes" on page 144

"Troubleshooting Cluster Database Recovery" on page 147

Prerequisites

To recover Windows 2000 or Windows 2003 MSCS host systems, ensure that each of the following prerequisites is satisfied:

• The NetWorker software is installed on each cluster node.

• Backups that include the SYSTEM or VSS SYSTEM save sets have been performed on a regular basis by a NetWorker server in the same domain as the cluster nodes. This helps to ensure that data is available for recovery to the desired point in time. The Microsoft cluster database is automatically included when the SYSTEM STATE or VSS SYSTEM

SERVICES save set is specified for backup or recovery.

• During a recovery, the domain controller for the domain to which the cluster nodes belong must be available to authenticate the node joining the cluster.

NetWorker V7.3

Disaster Recovery Guide

139

8

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

Recovering a Failed Quorum Disk

This section describes the procedure for recovering a failed quorum disk. The procedure requires the following:

• The quorum disk is designated for exclusive use by MSCS.

• The quorum disk resides in a drive array that is not a RAID.

• A spare disk, identical in type to the quorum disk, must be available to replace the failed disk. The spare disk must have at least as much total disk space as the quorum disk that is being replaced.

• The new quorum disk must be assigned the same drive letter as the failed disk.

For related information, refer to Microsoft Knowledge Base article Q280353:

How to Change Quorum Disk Designation.

How to Recover a Failed Quorum Disk

To recover a failed quorum disk:

1. Close all instances of Microsoft Cluster Administrator.

2. Stop the cluster service on both nodes.

3. Using the Computer Management Services tool, access the Cluster Service

Properties dialog box and change the startup type to Manual on both nodes.

4. Power down both nodes and the drive array.

5. Replace the failed quorum disk.

6. Power up both nodes and the drive array.

7. Format the new disk with the same partitioning scheme, disk format, drive letter, label, and mount points as the failed quorum disk. Verify that the new disk appears identical on both nodes.

8. On one node, use the Computer Management Services facility to access the

Cluster Service Properties dialog box.

9. Add -fixquorum as a start parameter and start the service.

10. Use the Cluster Administrator to rename the failed Quorum resource to

RemoveMe

.

140

NetWorker V7.3

Disaster Recovery Guide

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

11. Create a new disk resource named as follows:

Disk

x

where x is the drive letter of the old quorum disk.

12. Place the new disk resource in the Cluster Group.

13. Bring this disk resource online.

14. Right-click the cluster name and select Properties.

15. Click the Quorum tab and make the new drive the Quorum resource.

16. Using the Computer Management Services tool, access the Cluster Service

Properties dialog box and: a. Stop the Cluster service. b. Remove -fixquorum as a start parameter. c. Start the Cluster service.

17. Start the Cluster service on the other node.

18. Delete the RemoveMe resource.

19. Bring the cluster group online.

20. Using the Computer Management Services tool, access the Cluster Service

Properties dialog box and change the Cluster service startup type to

Automatic on both nodes.

21. Reboot both nodes.

Recovering One Cluster Node

This section describes how to recover the Cluster Server and cluster database if one of the cluster nodes fails.

Unexpected software behavior, such as corruption of a cluster-critical file, may result in a partitioned cluster. In such cases, one node of the cluster is unaware of the presence of the other operating node. Each node may attempt to take control of the shared quorum device, thus potentially rendering one node unable to function as a member of the cluster.

8

NetWorker V7.3

Disaster Recovery Guide

141

8

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

How to Recover One Cluster Node

The following procedure is based on a scenario in which Node_B has failed, and the Quorum resource has successfully failed over to Node_A. The operating system on each node is still intact, and therefore does not require recovery. The status of MSCS on Node_B is unknown, so it must be reinstalled

(if using Windows 2000) or cleaned up (if using Windows 2003).

To recover Node_B in this scenario:

1. Using the Cluster Administrator on Node_A, evict Node_B from the cluster.

2. On Node_B, reinstall or clean up MSCS as follows:

Windows 2000

: a. Using the Control Panel Add/Remove Programs tool, uninstall the

Windows component Cluster Service.

b. Reboot Node_B.

c. Log on to Node_B as Administrator for the domain in which the cluster nodes reside.

d. Using the Control Panel Add/Remove Programs tool, add the

Windows component Cluster Service.

Windows 2003:

At the command prompt, enter the following command to clean up MSCS:

cluster node

Node_B

/forcecleanup

Important:

After reinstalling the Cluster Service (Windows 2000) or cleaning up the Cluster Server (Windows 2003), you must reproduce the preexisting application environment on Node_B so MSCS can administer the applications as it did before the node failed. One way to reproduce the preexisting application environment is to perform disaster recovery procedures for each application that was previously installed on the node.

3. Using the Cluster Administrator, select Join An Existing Cluster and enter the cluster name.

4. Using the Cluster Administrator on Node_A, ensure that Node_B appears available for failover operations.

142

NetWorker V7.3

Disaster Recovery Guide

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

5. Only Node_A, the node that owns the shared resources, can be running while the cluster database is being recovered. Stop the Cluster service on

Node_B by using one of the following methods:

• Using the Control Panel Services tool, stop the Cluster service.

• At the command prompt, enter the following:

net stop clussvc

6. On Node_B, start the NetWorker User program and click Recover.

7. In the Source Client dialog box, select Node_B.

8. In the Destination Client dialog box, select Node_B.

9. Mark the following save sets for recovery.

• All local drives (except the NetWorker installation directory)

• SYSTEM FILES:

• SYSTEM DB:

• SYSTEM STATE:

Alternatively, if you are using Windows 2003 and Volume Shadow Copy

Service (VSS) is licensed and enabled, mark the following save sets:

• All local drives (except the NetWorker installation directory)

• VSS SYSTEM BOOT:

• VSS SYSTEM FILESET:

• VSS SYSTEM SERVICES:

• VSS USER DATA:

• VSS OTHER:

• VSS ASR DISK:

Note:

If you recover save sets in multiple NetWorker User operations, recover the SYSTEM STATE or VSS SYSTEM BOOT save set last to avoid booting Windows twice.

10. Click Start to begin the recovery.

11. After the recovery, reboot Node_B.

12. Using Cluster Administrator on Node_A, confirm that the cluster resources were recovered to the point in time when the backup occurred.

If you have been performing regular scheduled backups, this will recover the cluster database to a point in time shortly before the loss of Node_B.

8

NetWorker V7.3

Disaster Recovery Guide

143

8

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

13. Start the Cluster service on Node_B by using one of the following methods:

• Using the Control Panel Services tool, start the Cluster service.

• At the command prompt, enter the following:

net start clussvc

14. Using the Cluster Administrator on Node_A, monitor the cluster joining status of Node_B.

15. Using the Cluster Administrator on Node_B, verify that the cluster group can be moved between the nodes by right-clicking the group and selecting

Move group.

Recovering Multiple Cluster Nodes

This section provides general guidelines for performing a cluster recovery in case both cluster nodes fail. In this scenario, the operating system is unusable on each node. Therefore, this recovery procedure includes reinstallation of

Windows on each node, as well as NetWorker recovery of the cluster database.

Important:

Because cluster configurations vary, it is not possible to provide cluster disaster recovery procedures for every situation. Depending on the particular cluster configuration and the nature of the failure, it might be necessary to vary some of the procedures described in this section.

How to Recover Multiple Cluster Nodes

To perform a complete cluster recovery in the event that both cluster nodes,

Node_A and Node_B, have failed:

1. Ensure that the system disk of each node is configured to use the same partitioning scheme and drive letter that were in use prior to the loss of the node.

Note:

On each cluster node, the repaired or replaced system disk must have the same partitioning scheme, disk format, mount points, and drive letter assignments that were in use prior to the node failure. However, drive letter assignments of available volumes during Windows setup may cause the system disk drive letter to change. Therefore, prior to reinstalling the Windows operating system, disconnect the SCSI bus if the shared drives are being enumerated before the local drives.

144

NetWorker V7.3

Disaster Recovery Guide

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

2. Format and verify the system disk on each node.

3. On each node, reinstall Windows to the partition it occupied before the node failure. During Windows setup on each node, verify that the domain controller is available and that each potential node is able to join.

4. Delete the MSCS folder on the quorum disk.

5. On the quorum disk, run the chkdsk command.

6. For Windows 2000 systems, reinstall the Cluster Service on Node_A: a. While Node_B is still detached from the shared SCSI bus, and by using the Control Panel Add/Remove Programs tool, add the

Windows component Cluster Service.

The Cluster Setup Wizard appears. b. Using the Cluster Setup Wizard, enter the same configuration information that was in use prior to the failure of the cluster nodes

(including user account, IP addresses, and cluster name).

c. Reboot Node_A.

Note:

This step is not required for Windows 2003 systems because the

Cluster Service is installed automatically with Windows 2003.

7. For Windows 2000 systems, reinstall the Cluster Service on Node_B: a. On Node_B, use the Control Panel Add/Remove Programs tool to add the Windows component Cluster Service.

The Cluster Setup Wizard appears.

b. Using the Cluster Setup Wizard, select Join An Existing Cluster and enter the cluster name.

c. Reboot Node_B.

Note:

This step is not required for Windows 2003 systems because the

Cluster Service is installed automatically with Windows 2003.

8. On Node_B, enter the following at the command prompt.

net stop clussvc

9. Reinstall the NetWorker client software on Node_A.

10. On Node_A, start the NetWorker User program and click Recover.

11. In the Source Client dialog box, select Node_A.

12. In the Destination Client dialog box, select Node_A.

8

NetWorker V7.3

Disaster Recovery Guide

145

8

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

13. Mark the following save sets for recovery.

• All local drives (except the NetWorker installation directory)

• SYSTEM FILES:

• SYSTEM DB:

• SYSTEM STATE:

Alternatively, if you are using Windows 2003 and Volume Shadow Copy

Service (VSS) is licensed and enabled, mark the following save sets:

• All local drives (except the NetWorker installation directory)

• VSS SYSTEM BOOT:

• VSS SYSTEM FILESET:

• VSS SYSTEM SERVICES:

• VSS USER DATA:

• VSS OTHER:

• VSS ASR DISK:

Note:

If you recover save sets in multiple NetWorker User operations, recover the SYSTEM STATE or VSS SYSTEM BOOT save set last to avoid booting Windows twice.

14. Click Start to begin the recovery process.

15. After the recovery of Node_A is complete, reboot Node_A.

16. On Node_A, run the Cluster Administrator to confirm that the states of the cluster resources were recovered to the desired point in time.

17. Start the cluster service on Node_B by using one of the following methods:

• At the command prompt, enter the following:

net stop clussvc

• Using the Computer Management Services tool, right-click Cluster

Server and select Start.

18. From the Cluster Administrator on Node_B, verify that the cluster group can be moved between the nodes by right-clicking the group and selecting

Move group.

19. Reinstall the NetWorker client software on Node_B.

146

NetWorker V7.3

Disaster Recovery Guide

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

Troubleshooting Cluster Database Recovery

Failure cases for a cluster database recovery include the following:

1. The cluster service is not installed. In this case, you must install the cluster service before recovering the cluster database.

2. The node name has changed. In this case, you must change the node name back to its original setting prior to recovering the cluster database.

3. A shared disk owned by a resource group has been replaced. This means that this resource group will remain offline even after the cluster database recovery. If a shared disk owned by a resource group has been replaced, a separate recovery specific to the application that owns the disk may also be needed.

4. The partition layout of the quorum disk has changed since the backup. For recovery information in this case, see the section

"Cluster Recovery

Environment Variable Settings" on page 147 .

5. The drive letter of the quorum disk is already being used. For recovery

information in this case, see the section "Cluster Recovery Environment

Variable Settings" on page 147

.

Cluster Recovery Environment Variable Settings

If a cluster database recover failed because the partition layout of the quorum disk has changed since the backup, or the drive letter of the quorum disk is already being used, the recovery process can be repeated with the following environment variables defined as follows:

%NsrClusterForceRecover% should be set to 1.

%NsrClusterNewQuorumDrive% should be set to the drive letter of the new quorum disk (for example, R:).

There should only be one active node during the recovery. If other nodes are active and the environment variable %NsrClusterForceRecover% is not defined, then the recovery will fail. With %NsrClusterForceRecover% defined, the recover process will attempt to shut down the cluster service on the other active nodes.

It will also attempt to change the signature and drive letter (if

%NsrClusterNewQuorumDrive% is defined) of the current quorum disk even if its partition layout does not match the one described in the backup. However, the success of the recovery is not guaranteed even with

%NsrClusterForceRecover% defined.

8

NetWorker V7.3

Disaster Recovery Guide

147

8

Chapter 8: Microsoft Cluster Disaster Recovery (Windows 2000 and 2003)

148

NetWorker V7.3

Disaster Recovery Guide

Chapter 9: VERITAS Cluster Server

Disaster Recovery

This chapter explains how to recover one or more computers in a VERITAS

Cluster Server 3.5 environment.

This chapter includes the following sections:

"Prerequisites" on page 149

"Recovering a Single Cluster Node" on page 150

"Recovering an Entire Cluster" on page 151

Prerequisites

To help prepare for a disaster, record the cluster configuration information and keep it up to date. Cluster configuration information includes, but is not limited to, the following:

• Hardware setup

• Operating system

• Software and patches level

• Cluster name and ID number

• VERITAS disk groups and logical volumes

• License keys

• Virtual IP addresses used for highly available applications.

For more information about operating system information that must be recorded, see

"Reinstalling and Configuring the UNIX Operating System" on page 24 .

For more information about cluster configuration, refer to the VERITAS

Cluster Server documentation.

NetWorker V7.3

Disaster Recovery Guide

149

9

Chapter 9: VERITAS Cluster Server Disaster Recovery

Additionally, ensure that you regularly back up both physical and virtual

NetWorker clients. A physical NetWorker client backs up a physical node. A virtual NetWorker client backs up a highly available application. An example of an application that can be set up as a highly available application is a web server.

Recovering a Single Cluster Node

This section describes how to recover a single cluster node. This scenario assumes that the operating system on the damaged node must be reinstalled.

How to Recover a Single Cluster Node

To recover a single node:

1. Replace the damaged node and ensure that it is connected to the network.

2. On the recovered node, reinstall the following:

• Operating system software. For more information, see

"Reinstalling and Configuring the UNIX Operating System" on page 24

.

• VERITAS Volume Manager and VERITAS File System, if they were used.

• VERITAS Cluster Server software. Configure disk partitions and logical volumes exactly as they were configured before the disaster.

• Applications, such as web server software. Also, reinstall the

VERITAS Cluster Server agents for these applications.

Important:

Ensure that the cluster’s disk partitions, logical volumes, and service groups are configured exactly as they were configured before the disaster.

For detailed instructions on installing the listed VERITAS items, refer to the VERITAS Cluster Server Installation Guide.

3. Reinstall and configure the NetWorker software.

For detailed installation and configuration instructions, refer to the

NetWorker Administrator’s Guide.

150

NetWorker V7.3

Disaster Recovery Guide

Chapter 9: VERITAS Cluster Server Disaster Recovery

4. Recover the physical node's private data. Private data is data on the node's local storage that is not available to all nodes in the cluster.

For information about recovering physical NetWorker clients, refer to the

NetWorker Administrator's Guide.

Important:

Do not recover the VERITAS Volume Manager database from backups. This database has information that is applicable only to the hardware that was installed before the disaster.

5. Copy the VERITAS Cluster Server configuration files from an active

VERITAS Cluster Server node. The configuration files include, but are not limited to the following:

– /etc/llthosts

– /etc/llttab

– /etc/gabtab

– Files in the /etc/VRTSvcs/conf/config directory, such as main.cf and

types.cf.

For detailed instructions, refer to the VERITAS Cluster Server Installation

Guide.

6. Start the node as part of the cluster.

For detailed instructions, refer to the VERITAS Cluster Server Installation

Guide.

Recovering an Entire Cluster

This section describes how to recover an entire cluster.

How to Recover an Entire Cluster

To recover an entire cluster, complete the following tasks:

"Task 1: Replace the Damaged Hardware" on page 152

"Task 2: Reinstall the Software" on page 152

"Task 3: Recover the Cluster Configuration and Start the Cluster" on page 153

"Task 4: Recover the Data" on page 153

NetWorker V7.3

Disaster Recovery Guide

151

9

9

Chapter 9: VERITAS Cluster Server Disaster Recovery

Task 1: Replace the Damaged Hardware

To replace the damaged hardware:

1. Replace all damaged nodes.

2. Ensure the network connections are working.

Task 2: Reinstall the Software

To reinstall the software:

1. On the recovered node, reinstall the following:

• Operating system software. For more information, see

"Reinstalling and Configuring the UNIX Operating System" on page 24

.

• VERITAS Volume Manager and VERITAS File System, if they were used.

• VERITAS Cluster Server software. Configure disk partitions and logical volumes exactly as they were configured before the disaster.

• Applications, such as web server software. Also, reinstall the

VERITAS Cluster Server agents for these applications.

For detailed instructions on installing the listed VERITAS items, refer to the VERITAS Cluster Server Installation Guide.

Important:

Ensure that the cluster’s disk partitions, logical volumes, and service groups are configured exactly as they were configured before the disaster.

2. Reinstall and configure the NetWorker server software on all cluster nodes.

For detailed installation instructions, refer to the NetWorker Administrator’s

Guide.

152

NetWorker V7.3

Disaster Recovery Guide

Chapter 9: VERITAS Cluster Server Disaster Recovery

Task 3: Recover the Cluster Configuration and Start the Cluster

To recover the cluster configuration:

1. Recover each physical node's private data. Private data is data on the node's local storage that is not available to all nodes in the cluster.

For information about recovering physical NetWorker clients, refer to the

NetWorker Administrator's Guide.

2. Start the cluster software. For detailed instructions, refer to the VERITAS

Cluster Server Installation Guide.

Task 4: Recover the Data

To recover the highly available application data, recover the virtual NetWorker clients.

For information about recovering virtual NetWorker clients, refer to the

NetWorker Administrator’s Guide.

9

NetWorker V7.3

Disaster Recovery Guide

153

9

Chapter 9: VERITAS Cluster Server Disaster Recovery

154

NetWorker V7.3

Disaster Recovery Guide

Chapter 10: Sun Cluster Disaster Recovery

This chapter explains how to recover one or more computers in a Sun Cluster

3.x environment.

This chapter includes the following sections:

"Prerequisites" on page 155

"Recovering a Single Cluster Node" on page 155

"Recovering an Entire Cluster" on page 156

Prerequisites

To help prepare for a disaster, complete the Sun Installation and Planning

Configuration worksheets. These worksheets contain information that you can use to recover a cluster to its original configuration. These worksheets are located in the Sun Cluster 3.0 Release Notes.

Additionally, ensure that you regularly back up both physical and virtual

NetWorker clients. A physical NetWorker client backs up a physical node. A virtual NetWorker client backs up a highly available Sun Cluster resource application. Examples of applications that can be set up as highly available

Sun Cluster resources include a NetWorker server or a web server application.

Recovering a Single Cluster Node

This section describes how to recover a single cluster node. This scenario assumes that the operating system on the damaged node must be reinstalled.

NetWorker V7.3

Disaster Recovery Guide

155

10

Chapter 10: Sun Cluster Disaster Recovery

How to Recover a Single Cluster Node

To recover a single node:

1. Replace the damaged node and ensure that it is connected to the network.

For information about replacing the hardware and verifying network connections, refer to the Sun Cluster 3.x Hardware Guide and the server documentation.

2. Reinstall the operating system and the Sun Cluster software on the recovered node and reconfigure the node to join the cluster. For detailed instructions, refer to the Sun Cluster Installation Guide.

3. Reinstall the NetWorker software.

Important:

The installation steps you perform vary depending on whether the

NetWorker server is set up as a highly available Sun Cluster resource or the

NetWorker server is set up outside of the cluster. For detailed installation and configuration instructions for Sun Clusters, refer to the NetWorker

Administrator’s Guide.

4. Recover the physical node’s private data by using the nwrecover command. Private data is data that is not available to all nodes in the cluster. For information about recovering physical NetWorker clients, refer to the NetWorker Administrator’s Guide.

Recovering an Entire Cluster

This section describes how to recover an entire cluster.

How to Recover an Entire Cluster

To recover an entire cluster, complete the following tasks:

"Task 1: Replace the Damaged Hardware" on page 157

"Task 2: Reinstall the Software" on page 157

"Task 3: Recover the Data" on page 157

156

NetWorker V7.3

Disaster Recovery Guide

Chapter 10: Sun Cluster Disaster Recovery

Important:

To help ensure a smooth recovery process, have a copy of the Sun

Installation and Planning Configuration worksheets on hand. These worksheets are located in the Sun Cluster 3.0 Release Notes and should be completed before setting up the Sun Cluster.

Task 1: Replace the Damaged Hardware

Replace all damaged nodes and ensure that network connections are working.

For information about replacing the hardware and verifying network connections, refer to the Sun Cluster 3.x Hardware Guide and the server documentation.

Task 2: Reinstall the Software

To reinstall the software:

1. Reinstall and reconfigure the operating system and Sun Cluster software on all cluster nodes. For detailed instructions, refer to the Sun Cluster

Installation Guide.

Important:

Ensure that you reconfigure the cluster, including the disk device groups and resource groups, exactly as it was before the disaster.

2. Reinstall the NetWorker server software on all cluster nodes and configure the NetWorker server as a highly available Sun Cluster resource.

Important:

If the NetWorker server was not set up as a highly available Sun

Cluster resource, you need only reinstall the NetWorker client on all cluster nodes. For more information, refer to the NetWorker Administrator’s Guide.

10

Task 3: Recover the Data

This section describes how to recover the NetWorker server as well as the physical and virtual NetWorker clients.

NetWorker V7.3

Disaster Recovery Guide

157

10

Chapter 10: Sun Cluster Disaster Recovery

On the primary node for the NetWorker server:

1. Recover the highly available NetWorker server data as you would a stand-alone NetWorker server. For information about recovering a

NetWorker server, see

"Chapter 3: UNIX Disaster Recovery" on page 21 .

This step recovers the media database, resource database, and client file indexes, all of which are used to recover NetWorker clients.

2. Recover the physical and virtual NetWorker clients. For information about recovering physical and virtual NetWorker clients, refer to the NetWorker

Administrator’s Guide.

158

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery

This chapter explains how to recover one or more computers in an HP

TruCluster Server environment, releases 5.0A to 5.1B.

This chapter includes the following sections:

"Prerequisites" on page 159

"Recovering One or More Nodes in a Cluster" on page 160

"Recovering an Entire Cluster" on page 163

Prerequisites

Preparing for a disaster involves two major tasks, recording information and backing up the cluster. These tasks are described in the following sections:

"Record Information" on page 159

"Back Up the Cluster" on page 160

Record Information

This section describes the information required to recover an entire cluster.

Keep an up-to-date copy of the following information in a safe place:

• Network settings for each node in the cluster.

• Partition information for the disk drives in the system.

• Mappings for the managed disks and volumes. To save this information, open the System Reference Manual (SRM) console and display the output of the Show Devices option. Record this output on paper and store the paper in a safe place.

NetWorker V7.3

Disaster Recovery Guide

159

11

Chapter 11: HP TruCluster Disaster Recovery

• System topology information. Generate this information by entering the

hwmgr -view

command:

hwmgr -view hierarchy >

sys_topology_file

where sys_topology_file is a file to which you are saving output from the

hwmgr -view hierarchy

command.

• The device layer information. Generate this information by entering:

hwmgr -view devices >

device_layer

where device_layer is a file to which you are saving output from the

hwmgr -view devices

command.

Back Up the Cluster

To ensure that the cluster is disaster-ready, perform the following steps:

1. Back up the entire cluster.

a. Set up a NetWorker Client resource for the cluster’s default alias.

b. Add the hostname of each physical cluster node to the Client resource’s Remote Access list.

c. Specify a save set of All for the Client resource. A save set of All ensures that file systems from all nodes are backed up.

d. Create a NetWorker Group resource, add the Client resource to this group, and schedule the group for regular backups.

e. Perform at least one full backup.

2. (Optional) Back up the stand-alone HP Tru64 boot drive on the cluster’s primary node to a remote NetWorker server. You may need to mount the drive partitions temporarily and execute a manual save.

Recovering One or More Nodes in a Cluster

This section explains how to recover one or more physical nodes in a cluster. If

all of the physical nodes in a cluster have been destroyed, see "Recovering an

Entire Cluster" on page 163 .

Note:

To help ensure a successful recovery, have on hand the information

listed in the section titled, "Prerequisites" on page 159 . This information should

have been recorded prior to the disaster.

160

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery

How to Recover One or More Nodes in a Cluster

To recover one or more physical cluster nodes, complete the following tasks:

"Task 1: Replace the Damaged Hardware" on page 161

"Task 2: Verify Hardware Mappings" on page 161

"Task 3: Partition the Down Node’s Boot Drive" on page 161

"Task 4: Update the Cluster Configuration" on page 161

"Task 5: Recover the Node-Specific Data" on page 162

Task 1: Replace the Damaged Hardware

Replace the damaged equipment with the same or similar model equipment as installed previously. Connect all the equipment as previously connected.

Task 2: Verify Hardware Mappings

Open the SRM console on the node you are recovering (down node) and set the logical mappings for the node’s managed disks and volumes to their original settings. To help complete this step, use the information that was recorded from the SRM console prior to the disaster.

Task 3: Partition the Down Node’s Boot Drive

To partition the down node’s boot drive, you need the original node’s partition information. This information should have been recorded prior to the disaster.

Note:

Perform this task from a working cluster node.

To partition the down node’s boot drive, use a utility such as diskconfig and specify the original partition settings.

Task 4: Update the Cluster Configuration

To update the cluster configuration, delete the cluster node, then add it back into the cluster. When you add the node back, cluster hardware settings are reinitialized.

Note:

Perform this task from a working cluster node.

11

NetWorker V7.3

Disaster Recovery Guide

161

11

Chapter 11: HP TruCluster Disaster Recovery

To update the cluster configuration:

1. Delete the cluster node:

clu_delete_member -m

member#

where member# is the number of the cluster node.

2. Remove the node’s boot partition domain:

rmfdmn root

member#

_domain

3. Add the computer back into the cluster by entering the clu_add_member command. For more information about adding a cluster node, refer to the

TruCluster documentation.

Task 5: Recover the Node-Specific Data

Once the cluster node has been added and configured, recover its node-specific data. Node-specific data includes information, such as node licenses and volume definitions that would be difficult and time-consuming to replace unless they are recovered.

Note:

Perform this task on the down node:

To recover the node-specific data:

1. Boot the cluster node.

boot -file genvmunix

dkc2

where dkc2 is the cluster disk.

2. Use the NetWorker recover command in interactive mode to recover the cluster node data, as shown in the following example.

Note:

The force option is used in this step so that the recovered files overwrite the existing files.

recover -c

cluster_default_alias

recover> add /usr/cluster/members/member

member#

recover> add /var/cluster/members/member

member#

recover> add /cluster/members/member

member#

/etc

recover> cd /cluster/members/member

member#

recover> delete ddr* recover> delete gen* recover> delete dfsl*

162

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery recover> delete dvrdevtab* recover> force recover> recover where cluster_default_alias is the name of the cluster and member# is the member ID of the down cluster node.

3. Use the NetWorker recover command to recover the node’s dev and devices directories, as in the following example.

Note:

The noforce option is used in this step so that the recovered files do

not overwrite the existing files.

recover -c

cluster_default_alias

recover> add /dev /devices recover> cd /cluster/members/member

member#

recover> add dev devices recover> noforce recover> recover where cluster_default_alias is the name of the cluster and member# is the member ID of the down cluster node.

Note:

When prompted to overwrite files, enter N to prevent files from being overwritten.

4. Recompile the kernel by using the doconfig command.

Recovering an Entire Cluster

This section explains how to recover an entire cluster that has been destroyed.

Note:

To help ensure a successful recovery, have on hand the information

listed in the section titled, "Record Information" on page 159 . This information

should have been recorded prior to the disaster.

How to Recover an Entire Cluster

To recover an entire cluster, perform the following tasks:

"Task 1: Replace the Damaged Hardware" on page 164

"Task 2: Verify Hardware Mappings" on page 164

"Task 3: Reinstall and Configure the Cluster Software" on page 164

NetWorker V7.3

Disaster Recovery Guide

163

11

11

Chapter 11: HP TruCluster Disaster Recovery

"Task 4: Reinstall the NetWorker Software" on page 165

"Task 5: Configure NetWorker Device and Autochanger Resources" on page 166

"Task 6: Locate the Server’s Bootstrap Save Set ID" on page 166

"Task 7: Recover the Server’s Bootstrap" on page 167

"Task 8: Locate the Server’s Client File Index Save Set IDs" on page 169

"Task 9: Recover the Server’s Client File Index" on page 169

"Task 10: Re-create the Cluster" on page 170

"Task 11: Recover the Cluster Data" on page 170

"Task 12: Create a Record of Backups Done after Bootstrap Created" on page 173

Task 1: Replace the Damaged Hardware

Replace the damaged equipment with the same or similar model equipment as installed previously. Connect all the equipment as previously connected.

Task 2: Verify Hardware Mappings

Open the SRM console and set the logical mappings for each node’s managed disks and volumes to their original settings. To help complete this step, use the information that was recorded from the SRM console prior to the disaster.

Task 3: Reinstall and Configure the Cluster Software

Perform a stand-alone installation of Tru64 on the primary node’s stand-alone boot drive. This task requires the cluster information that was recorded prior to the disaster.

To reinstall and configure the software:

1. Reinstall the base operating system and TruCluster package on the primary node’s stand-alone boot drive. For more information, refer to the

TruCluster installation documentation.

2. Configure the network resources.

3. Verify that the system topology matches what you recorded, and the device layer is configured the same, as follows:

hwmgr -view hierarchy hwmgr -view devices

164

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery

4. If the logical disks are enumerated differently than in the original configuration, manually reapply the original enumeration to the disks.

If you need to remap device numbers, enter the -e option with the following command:

dsfmgr -e

dsk10 dsk4

where dsk10 is the new disk enumeration value and dsk4 is the original disk enumeration value.

If you want to move a device number, enter the -m option.

dsfmgr -m

dsk10 dsk4

Task 4: Reinstall the NetWorker Software

Perform this task if the NetWorker server was installed as a highly available cluster application. If the NetWorker server is outside the cluster, skip this task and proceed to

"Task 9: Recover the Server’s Client File Index" on page 169

.

To reinstall NetWorker software:

1. Reinstall the same version of the NetWorker server software. When prompted during the installation, enter the location of the /nsr directory as follows:

/nsr

For more information about installing NetWorker software, refer to the platform appropriate EMC NetWorker Installation Guide.

2. Reinstall any NetWorker patches that were installed prior to the disaster.

3. If separate file systems were created for the subdirectories under the original /nsr directory, mount those file systems under this /nsr directory.

Separate file systems may have been created for subdirectories, such as

/nsr/mm or /nsr/index, that became very large.

4. Re-create the cluster’s /nsr directory:

mkdir -p /

cluster_default_alias

/nsr cd /cluster/members/member0 ln -s /

cluster_default_alias

/nsr nsr cd /cluster/members/member1 ln -s /

cluster_default_alias

/nsr nsr

11

NetWorker V7.3

Disaster Recovery Guide

165

11

Chapter 11: HP TruCluster Disaster Recovery

Task 5: Configure NetWorker Device and Autochanger Resources

To configure the Device and Autochanger resources:

• To recover data by using a stand-alone device, ensure that the stand-alone

Device resource exists (defined in the /nsr/res directory). If the stand-alone

Device resource does not exist, create it by using the NetWorker

Administrator program.

• To recover data with an autochanger, ensure that the Autochanger resource exists (defined in the /nsr/res directory). If the Autochanger resource does not exist, create it using the NetWorker Console

Administration interface. For details about the NetWorker Console

Administration interface, refer to the NetWorker Administrator’s Guide.

– If you are using an autochanger, reset the autochanger by entering the

nsrjb -vHE

command. This command resets the autochanger, ejects backup volumes, reinitializes the element status, and checks each slot for a volume.

If the autochanger does not support the -E option, initialize the element status by using sjiielm.

– If you are using an autochanger, inventory the autochanger by entering the nsrjb -I command. This helps determine whether the volumes required to recover the bootstrap are located inside the autochanger.

Task 6: Locate the Server’s Bootstrap Save Set ID

Perform this task if you need to locate the bootstrap save set ID in order to recover the NetWorker server’s latest bootstrap. If you have a record of the save set ID for the bootstrap, proceed to

"Task 7: Recover the Server’s

Bootstrap" on page 167 .

Important:

If you move the NetWorker backup media to an offsite location and a subsequent file recover operation generates a mount request, the recover operation will wait until an operator satisfies the mount request. To avoid delays, use the mminfo -mv command to list the media associated with the file you are recovering. If necessary, retrieve the media from the offsite storage before starting the recover.

166

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery

To locate the most recent bootstrap save set ID:

1. Insert the most recent media or clone volumes used for scheduled backups into the appropriate device.

2. At the system prompt, switch to the directory where the NetWorker binaries and executables are located.

3. If you are using an autochanger, insert the first volume of the bootstrap save set into the first drive of the autochanger by entering the following command:

nsrjb -lnv -S

slot

-f

device-name

where:

slot is the slot where the first volume is located

device-name is the pathname for the first drive. You can obtain this

device-name by using the inquire command.

4. Determine the save set ID of the most recent bootstrap on the media, as follows:

scanner -B

device_name

where device_name identifies a device such as /dev/ntape/tape0_d1.

If you do not locate the save set ID of the most recent bootstrap on the most recent media, run the scanner -B command on preceding media to locate the save set ID of the most recent bootstrap.

5. When you see the output, record both the bootstrap save set ID and the volume label.

Task 7: Recover the Server’s Bootstrap

To recover the NetWorker server bootstrap:

1. Enter the mmrecov command to recover the NetWorker server’s bootstrap

(media database and resource configuration files):

mmrecov

11

NetWorker V7.3

Disaster Recovery Guide

167

11

Chapter 11: HP TruCluster Disaster Recovery

168

Important:

The mmrecov command overwrites the server’s media database. It

does not overwrite the resource configuration files, but instead mmrecov recovers them to the resource directory, res.R. For syntax and option information about mmrecov, refer to the mmrecov man page or the NetWorker

Command Reference Guide.

2. If the server has multiple devices configured and enabled, enter the name of the device you are using for the recovery when the following message appears:

What is the name of the device you plan on using

/dev/ntape/tape0_d1

3. When the following message appears, enter the save set ID for the latest bootstrap. If you are recovering a cloned version of the bootstrap, specify the save set ID associated with the clone.

Enter the latest bootstrap save set ID []:

20076

4. When the following message appears, enter the file number to begin the recovery. If unknown, press [Enter].

Enter starting file number (if known) [0]:

130

5. When the following message appears, enter the first record number to begin the recovery. If unknown, press [Enter].

Enter starting record number (if known) [0]:

0

6. When the following message appears, respond to the prompt:

Please insert the volume on which save set id 20076 started into /dev/ntape/tape0_d1.

Once you have loaded the appropriate volume, the following message appears:

Scanning /dev/ntape/tape0_d1 for save set 20076; this might take a while...

NetWorker software then scans the volume for the appropriate save set and recovers it. The NetWorker media database and resource configuration files are recovered when the following message appears:

If your resource files were lost, they are now recovered in the ‘res.R’ directory. Copy or move them to the ‘res’ directory, after you have shut down the service. Then

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery restart the service.

Otherwise, just restart the service.

If the on-line index for the server-name was lost, it can be recovered using the nsrck command.

Task 8: Locate the Server’s Client File Index Save Set IDs

Locate the save set IDs of the NetWorker server’s client file index. If there have been incremental backups since the last full backup, you need the save set ID of the last full backup and the save set IDs of each incremental backup since the last full backup.

To locate the client file index save set IDs, enter the mminfo command to display the client file index’s save set ID:

mminfo -av -N

index:

cluster_default_alias

where cluster_default_alias is the name of the cluster.

Task 9: Recover the Server’s Client File Index

In this task, you recover the most recent client file index from the last full backup and all subsequent incremental backups. The client file index contains detailed information about the backups performed for each NetWorker client.

1. Use the recover command to recover the client file indexes. Recover the most recent full backup first, then recover the subsequent incremental backups from earliest to latest. For example, if the dates for the last full backup and the last two incremental backups are as follows:

• Oct. 15, last full backup

• Oct. 16, second-to-last incremental backup

• Oct. 17, last incremental backup

Then, to recover the client file indexes, enter the following:

recover -S

full_SSID

-iY recover -S

second_last_incremental_SSID

-iY recover -S

last_incremental_SSID

-iY

where the _SSID variables are the save set IDs of the last full backup, the second-to-last incremental backup, and the last incremental backup, respectively.

11

NetWorker V7.3

Disaster Recovery Guide

169

11

Chapter 11: HP TruCluster Disaster Recovery

2. Enter the nsrim command to crosscheck the media database and index database:

nsrim -X

3 . Enter the nsrck command to check and repair the client file indexes if necessary:

nsrck -L6

4. Verify that the indexes were recovered properly:

nsrinfo

cluster_default_alias

where cluster_default_alias is the name of the cluster.

Task 10: Re-create the Cluster

To re-create the cluster, run the clu_create command:

clu_create

For more information about installing the cluster, refer to the TruCluster documentation.

Note:

Do not reboot after running clu_create. Remain in stand-alone HP Tru64 mode.

Task 11: Recover the Cluster Data

To recover the cluster data:

1. Mount the root, usr, and var cluster directories:

mount

cluster_root#root

/mnt

mount

cluster_usr#usr

/mnt

/usr mount

cluster_var#var

/mnt

/var

where /mnt is the mount point.

170

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery

2. Recover the cluster data by using the recover command in interactive mode, as done in the following example.

recover -c

cluster_default_alias

recover> cd / recover> add .* recover> add * recover> delete /devices /dev recover> cd /etc recover> delete dec* dccd* dcdd* ddr* recover> delete disktab dvrdevtab gen* dfsc* recover> cd /cluster/members/member0 recover> delete boot_partition dev devices recover> delete etc/ddr* recover> delete etc/gen* recover> delete etc/dfsl* recover> delete dvrdevtab recover> cd /cluster/members/member1 recover> delete boot_partition dev devices recover> delete etc/ddr* recover> delete etc/gen* recover> delete etc/dfsl* recover> delete dvrdevtab where cluster_default_alias is the name of the cluster.

11

NetWorker V7.3

Disaster Recovery Guide

171

11

Chapter 11: HP TruCluster Disaster Recovery

3. Repeat the following delete command for each cluster member, except for cluster members 0 and 1.

Note:

The force option is used in this step so that the recovered files overwrite the existing files.

recover> delete /cluster/members/member

member#

recover> force recover> relocate /mnt recover> recover where member# is the member ID of any cluster member except node 0 or 1.

4. Use the recover command in interactive mode to recover the dev and

devices directories for each cluster node as in the following example.

Note:

The noforce option is used in this step so that recovered files do not overwrite existing files.

recover -c

cluster_default_alias

recover> add /dev /devices recover> cd /cluster/members/member0 recover> add dev devices recover> cd /cluster/members/member1 recover> add dev devices recover> noforce recover> relocate /mnt recover> recover where cluster_default_alias is the name of the cluster.

Note:

When prompted to overwrite files, enter N.

5. Unmount the /mnt directory:

unmount

/mnt/var /mnt/usr /mnt

6. Shut down the system and boot the primary node off of its shared

HP Tru64 member boot drive.

7. (Optional) Recompile the kernel by using doconfig on the primary node, and then reboot.

172

NetWorker V7.3

Disaster Recovery Guide

Chapter 11: HP TruCluster Disaster Recovery

Task 12: Create a Record of Backups Done after Bootstrap Created

After a disaster recovery, the media database will not contain information about any backups done after the bootstrap was created. The media database retains information about the last known file/record marks on the tape, and if backups were done after the bootstrap was created, the file/record information will not be current. For any tapes that contain backups that were completed after the bootstrap was created, use scanner with, at a minimum, the -m option to scan the save sets and/or tapes into the media database. This will ensure that the media database has a record of all completed save sets and that valid backups will not be overwritten by future backups.

Additionally, write protect all appendable volumes until you can complete the scanner activity to prevent scheduled backups from using these volumes. See

"Task 8: Create a Record of Backups Done after Bootstrap Created" on page 37

for instructions on write protecting and scanning.

For information about recovering the remaining cluster nodes, see

"Recovering One or More Nodes in a Cluster" on page 160 .

11

NetWorker V7.3

Disaster Recovery Guide

173

11

Chapter 11: HP TruCluster Disaster Recovery

174

NetWorker V7.3

Disaster Recovery Guide

Chapter 12: HP-UX Cluster Disaster Recovery

This chapter explains how to recover one or more computers in an HP-UX

MC/ServiceGuard cluster environment.

This chapter includes the following sections:

"Recovering a Single Cluster Node" on page 175

"Performing a Cluster-Wide Recovery" on page 176

Recovering a Single Cluster Node

This section describes how to recover a cluster server on one node. This scenario assumes that the operating system on the participating nodes has failed and must be reinstalled. The functionality of the MC/ServiceGuard component on the node is also in question.

How to Recover a Single Cluster Node

To recover a single node:

1. Reinstall the operating system and the cluster software. Configure the volume groups.

2. Add the node to the cluster: a. Click Cluster in the System Administration Manager (SAM).

Note:

You can also use the cmrunnode command instead of SAM.

b. Select High Availability Clusters>Cluster Administration>Specify

Nodes to Join Cluster.

3. Recover the node’s data from a recent backup.

NetWorker V7.3

Disaster Recovery Guide

175

12

Chapter 12: HP-UX Cluster Disaster Recovery

Performing a Cluster-Wide Recovery

This section describes how to perform a complete cluster recovery, including reinstalling the operating system on all nodes in the cluster.

How to Perform a Cluster-Wide Recovery

To recover an entire cluster, perform the following tasks:

"Task 1: Reinstall the Software" on page 176

"Task 2: Recover the Cluster Database" on page 176

"Task 3: Start the Cluster" on page 177

"Task 4: Recover the Data" on page 178

Task 1: Reinstall the Software

To reinstall the software:

1. Reinstall the operating system and cluster software on each node.

2. Reinstall the NetWorker software as required within the HP-UX

MC/ServiceGuard cluster environment. For detailed instructions, refer to the EMC NetWorker Installation Guide, HP-UX Version.

Task 2: Recover the Cluster Database

1. Select any node in the cluster and run the NetWorker server and client software on that node.

2. Re-create the cluster configuration and the NetWorker software configuration and control files. The default location for the cluster configuration file is /etc/cmcluster and the default directory for the

NetWorker configuration and control files is /etc/cmcluster/NetWorker.

• If you know where the latest save sets are located, enter the scanner command to recover the configuration files:

scanner -S saveset_id

device_path

| uasm -i y -rv

where device_path is the path to the tape device.

• If you do not know the save set ID, enter the scanner device_path command to determine the saveset ID, as follows:

scanner

device_path

> /tmp/savefile 2>&1

176

NetWorker V7.3

Disaster Recovery Guide

Chapter 12: HP-UX Cluster Disaster Recovery

3. Run the NetWorker NetWorker.cluster script located in the

/opt/NetWorker/bin/ directory.

Note:

Do not re-create the legato.control and pkg.control files when prompted by the NetWorker.cluster script.

4. If the shared disk has to be replaced, configure the disk and file system of the replacement disk.

5. Create the file system, as follows:

vgchange -c n /dev/vg03 vgchange -a y /dev/vg03 newfs -F hfs /dev/vg03/rlvol1 vgchange -a n /dev/vg03

6. Enter the following command to delete the existing cluster and package configurations:

cmdeleteconf -c cluster1

7. Enter the following command to apply the cluster and package configurations, including the NetWorker package, to all the nodes within the cluster:

cd /etc/cmcluster cmapplycomf -C cluster1.ascii -P networker/pkg.conf \

-P other_pkg/other_pkg

8. Run the cmgetconf command to save the new cluster or package ASCII configuration file whenever the cluster or package configuration is created or modified, as follows:

cmgetconf -c cluster1 cluster1.ascii

cmgetconf -p networker pkg.conf

These ASCII files can be saved in the /etc/cmcluster directory of a node.

Ensure that the NetWorker software backs up /etc/cmcluster so that the configurations can be recovered by using cmapplyconf whenever necessary.

Task 3: Start the Cluster

To start the cluster, enter the following command:

cmruncl

This also starts the NetWorker software.

NetWorker V7.3

Disaster Recovery Guide

177

12

12

Chapter 12: HP-UX Cluster Disaster Recovery

Task 4: Recover the Data

To recover the data:

1. Recover the highly available NetWorker server data as you would a stand-alone NetWorker server. For information about recovering a

NetWorker server, see

"Chapter 3: UNIX Disaster Recovery" on page 21 .

When you shut down the NetWorker software during this step, use

cmhaltpkg

and not nsr_shutdown, as follows:

cmhaltpkg networker exchange -a c /dev/vg03 mount /dev/vg03/lvol1 /vg031 mv /vg031/nsr/res /vg031/nsr/res.old mv /vg031/nsr/res.R /vg031/nsr/res umount /vg031 vgchange -a n /dev/vg03 cmmodpkg -e networker

(restarts NetWorker software)

2. Recover the clients' data from recent backups.

178

NetWorker V7.3

Disaster Recovery Guide

Chapter 13: EMC FullTime AutoStart for AIX,

HP-UX, Linux, and Solaris Disaster Recovery

This chapter explains how to recover one or more computers in a EMC

Fulltime Autostart for an AIX, HP-UX, Linux, or Solaris environment.

This chapter includes the following sections:

"Recovering a Single Cluster Node" on page 179

"Performing a Cluster-Wide Recovery" on page 181

"Renaming the Resource Directory" on page 186

Recovering a Single Cluster Node

This section describes how to recover a failed node within a cluster. This scenario assumes that the operating system on the participating nodes is intact and will not be recovered. The functionality of the Fulltime Autostart component on the node is in question.

How to Recover a Single Cluster Node

To recover a single node:

1. Reinstall the NetWorker software. For detailed installation and configuration instructions, refer to the platform-appropriate NetWorker

Installation Guide or, if using a Fujitsu Siemens Computers NetWorker host, the NetWorker Administrator’s Guide.

2. Start the NetWorker client.

3. Using the cluster’s virtual NetWorker server running on an unaffected node, recover the node's data and software from a recent backup. For detailed instructions, refer to the NetWorker Administrator’s Guide.

NetWorker V7.3

Disaster Recovery Guide

179

13

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

180

4. Make sure to recover the following directories:

AIX:

• /etc/rc*

• $FT_DIR/config/*

• $FT_DIR /log/*

• $FT_DIR/<domain_name><node_name>

HP-UX:

• /etc/rc*

• $FT_DIR/config/*

• $FT_DIR /log/*

• $FT_DIR/<domain_name><node_name>

Solaris and Linux:

• /etc/rc*.d

• /etc/init.d

• $FT_DIR/config/*

• $FT_DIR /log/*

• $FT_DIR/<domain_name><node_name>

5. Recover or copy the following files from the active node:

AIX:

$FT_DIR/bin/envsh ($FT_DIR/bin/envcsh if using C shell)

• /usr/bin/nw_ux.lc

HP-UX:

$FT_DIR/bin/envsh ($FT_DIR/bin/envcsh if using C shell)

• /opt/networker/bin/nw_ux.lc

Solaris and Linux:

$FT_DIR/bin/envsh ($FT_DIR/bin/envcsh if using C shell)

• /usr/sbin/nw_ux.lc

6. Start the cluster virtual server on the node, as follows:

cd $FT_DIR/bin

(or source

./envcsh

if using C shell)

. ./envsh

./ft_startup -domain

domain_name

NetWorker V7.3

Disaster Recovery Guide

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

If the virtual server fails to start on the node because of an outdated sites file, copy the following file from the active node:

$FT_DIR/config/domain_name_sites

7. Run the networker.cluster script, as follows:

AIX:

/usr/bin/networker.cluster

HP-UX:

/opt/networker/bin/networker.cluster

Solaris and Linux:

/usr/sbin/networker.cluster

If you have already recovered or copied the nw_ux.lc file from an active node, enter No at the following prompt:

Do you wish to automatically add site-specific values for:

NSR_SHARED_DISK_DIR and NSR_SERVICE_ID in

/Fulltime_Autostart_ installation_ directory

Yes or No [Yes]? no

Performing a Cluster-Wide Recovery

This section describes how to perform a complete cluster recovery, including reinstalling the operating system on all nodes in the cluster.

How to Perform a Cluster-Wide Recovery

Perform the following tasks for a cluster-wide recovery:

"Task 1: Reinstall the Software" on page 182

"Task 2: Recover the Cluster" on page 182

"Task 3: Configure NetWorker Software as a Highly Available

Application" on page 183

"Task 4: Recover the Data" on page 184

"Task 5: Start the Cluster Software on Each Node" on page 185

"Task 6: Create a Record of Backups Done after Bootstrap Created" on page 186

This procedure assumes a worst-case scenario, where one node of a cluster with two or more nodes loses contact and completely fails, followed shortly by the remaining nodes completely failing.

13

NetWorker V7.3

Disaster Recovery Guide

181

13

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

Task 1: Reinstall the Software

To reinstall the software, on a primary node:

1. Configure the hardware and reinstall the operating system.

2. Reinstall the Fulltime Autostart software. For detailed instructions, refer to the EMC Automated Availability Manager Installation Guide.

3. Reinstall the NetWorker software. For detailed instructions, refer to the platform-appropriate NetWorker Installation Guide or, if using a Fujitsu

Siemens Computers NetWorker host, the NetWorker Administrator’s Guide.

4. Make sure that the NetWorker client and server software are running locally.

Task 2: Recover the Cluster

1. Determine the save set ID that contains the following cluster configuration and system files:

Note:

If you do not know the save set ID, enter the scanner device_path command to determine its value, as follows:

scanner

device_path

>

/tmp/savefile

2>&1

Note:

In the following lists, $FT_DIR, is the installation location for the

Fulltime Autostart software.

AIX:

• /usr/bin/nw_ux.lc

• $FT_DIR/config/*

• $FT_DIR /log/*

• $FT_DIR/<domain_name><node_name>

$FT_DIR/bin/envsh (envcsh for C-shell)

HP-UX:

• /opt/networker/bin/nw_ux.lc

• $FT_DIR/config/*

• $FT_DIR /log/*

• $FT_DIR/<domain_name><node_name>

$FT_DIR/bin/envsh (envcsh for C-shell)

182

NetWorker V7.3

Disaster Recovery Guide

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

Solaris and Linux:

• /usr/sbin/nw_ux.lc

• $FT_DIR/config/*

• $FT_DIR /log/*

• $FT_DIR/<domain_name><node_name>

$FT_DIR/bin/envsh (envcsh for C-shell)

2. On the primary node, enter the scanner command to recover the following configuration files:

scanner -S

saveset_id device_path

-x uasm -rv -i Y path_name

where the device_path is the platform configuration and system files listed in step 1.

Task 3: Configure NetWorker Software as a Highly Available Application

On the primary node, configure the NetWorker software as a highly available application:

1. Run the envsh script. For example:

AIX: cd /usr/lpp/LGTOaam

xx

/bin

. ./envsh

(or source

./envcsh

if using C shell) where xx is the version of Fulltime Autostart.

HP-UX: cd /opt/LGTOaam

xx

/bin

. ./envsh

(or source

./envcsh

if using C shell) where xx is the version of Fulltime Autostart.

Solaris and Linux: cd /opt/LGTOaam

xx

/bin

. ./envsh

(or source

./envcsh

if using C shell) where xx is the version of Fulltime Autostart.

13

NetWorker V7.3

Disaster Recovery Guide

183

13

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

2. Run the networker.cluster script, and if you have already recovered the

nw_ux.lc file, enter No at the prompt:

Do you wish to automatically add site-specific values for:

NSR_SHARED_DISK_DIR and NSR_SERVICE_ID in

/Fulltime_Autostart_ installation_ directory

Yes or No [Yes]? no

3. Start the cluster virtual server, as follows:

$FT_DIR/bin/ft_startup -domain

domain_name

4. Use the Fulltime Autostart Console to start the NetWorker resource group.

Task 4: Recover the Data

The procedures to recover the data on the primary and each of the remaining nodes in the cluster differ. For more information, see the following sections:

"On the Primary Node" on page 184

"On Each Remaining Node in the Cluster" on page 185

On the Primary Node

To recover data on the primary node:

1. Enter the mmrecov command to recover the indexes, media database, and resource files.

2. Enter the recover command to recover the following initialization directories:

AIX:

• /etc/rc*

HP-UX:

• /etc/rc*

Solaris and Linux:

• /etc/init.d

• /etc/rc*.d

For detailed instructions on using the recover and mmrecov commands, refer to the NetWorker Command Reference Guide.

184

NetWorker V7.3

Disaster Recovery Guide

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

On Each Remaining Node in the Cluster

To recover data on each remaining node in the cluster:

1. Reinstall NetWorker software and run the NetWorker client.

2. Recover the data and software from recent backups or from the product installation CD-ROMs.

3. Using the virtual NetWorker server, recover the following files:

• $FT_DIR/config/*

• $FT_DIR /log/*

• $FT_DIR/<domain_name>_<node_name>

4. Recover or copy the following files from the active node:

AIX:

• /usr/bin/nw_ux.lc

$FT_DIR/bin/envsh (envcsh for C-shell)

HP-UX:

• /opt/networker/bin/nw_ux.lc

$FT_DIR/bin/envsh (envcsh for C-shell)

Solaris and Linux:

• /usr/sbin/nw_ux.lc

$FT_DIR/bin/envsh (envcsh for C-shell)

For detailed instructions on using the recover and mmrecov commands, refer to the NetWorker Command Reference Guide.

Task 5: Start the Cluster Software on Each Node

On each of the remaining nodes in the cluster:

1. Start the cluster virtual server, as follows:

cd $FT_DIR/bin

. ./envsh

./ft_startup -domain

domain_name

If the virtual server fails to start on the node because of an outdated sites file, copy the following file from the active node:

$FT_DIR/config/

domain_name_sites

NetWorker V7.3

Disaster Recovery Guide

185

13

13

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

2. Run the networker.cluster script and if you have already recovered or copied the nw_ux.lc file from the active node in step 3, enter No at the prompt:

Do you wish to add site-specific values for:

NSR_SHARED_DISK_DIR and NSR_SERVICE_ID in

/Fulltime_Autostart_ installation_ directory

Yes or No [Yes]? no

3. Repeat steps 1 and 2 for each node in the cluster.

Task 6: Create a Record of Backups Done after Bootstrap Created

After a disaster recovery, the media database will not contain information about any backups done after the bootstrap was created. The media database retains information about the last known file/record marks on the tape, and if backups were done after the bootstrap was created, the file/record information will not be current. For any tapes that contain backups that were completed after the bootstrap was created, use scanner with, at a minimum, the -m option to scan the save sets and/or tapes into the media database. This will ensure that the media database has a record of all completed save sets and that valid backups will not be overwritten by future backups.

Additionally, write protect all appendable volumes until you can complete the scanner activity to prevent scheduled backups from using these volumes. See

"Task 8: Create a Record of Backups Done after Bootstrap Created" on page 37

for instructions on write protecting and scanning.

Renaming the Resource Directory

This section describes how to rename the resource directory. While performing a disaster recovery on a shared disk, you may need to rename the resource directory.

To rename the resource directory:

1. Use the Fulltime Autostart Console to stop the NetWorker resource group.

2. Use the Fulltime Autostart Console to activate and mount the following: a. If using the Logical Volume Manager, activate the volume group by bringing the logical volume group datasource online.

b. Mount the shared disk (that contains the shared nsr directory) by bringing the shared disk datasource online.

186

NetWorker V7.3

Disaster Recovery Guide

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

Note:

You can also activate and mount the logical volume group and the shared disk by using UNIX commands.

3. Move the newly recovered resource directory nsr/res.R to the current resource directory, as follows:

mv /nsr_shared_mnt_pt/nsr/res /nsr_shared_mnt_pt/ nsr/res.old

mv /nsr_shared_mnt_pt/nsr/res.R /nsr_shared_mnt_pt/ nsr/res

4. In the reverse order, take offline the data sources (logical volume group and the shared disk) that you used in step 2.

5. Use the Fulltime Autostart Console to start the NetWorker resource group.

NetWorker V7.3

Disaster Recovery Guide

187

13

Chapter 13: EMC FullTime AutoStart for AIX, HP-UX, Linux, and Solaris Disaster Recovery

13

188

NetWorker V7.3

Disaster Recovery Guide

Chapter 14: EMC FullTime AutoStart for

Windows Disaster Recovery

This chapter explains how to recover a cluster node in a EMC FullTime

AutoStart cluster for Windows 2000.

This chapter includes the following sections:

"Prerequisites" on page 189

"Recovering a Single Cluster Node" on page 189

Prerequisites

To recover a node in a FullTime AutoStart cluster, ensure that the following prerequisites are satisfied:

• The local drives on each cluster node are backed up on a regular basis.

These backups must occur on a NetWorker server that is outside the cluster.

• The NetWorker server is functioning and available on the network.

Recovering a Single Cluster Node

This section describes how to recover a failed node within a cluster. In this procedure’s scenario, the failed node is Node_B.

To recover a single cluster node:

1. Reinstall the operating system. Ensure that it is configured identically to the way it was configured before Node_B failed. Use the same drive letters for the disk volumes as those that were used prior to the failure.

2. Install and configure backup devices.

NetWorker V7.3

Disaster Recovery Guide

189

Chapter 14: EMC FullTime AutoStart for Windows Disaster Recovery

3. Reinstall the NetWorker client software on Node_B.

For detailed installation and configuration instructions, refer to the EMC

NetWorker Installation Guide, Microsoft Windows Version.

4. Start the NetWorker User program.

5. Select, mark, and recover all items on the node, including the FullTime

AutoStart software and data. Be sure to select and mark the cluster’s shared drives so that they are included in the recovery operation.

For detailed instructions about performing recovery operations, refer to the NetWorker Administrator’s Guide.

14

190

NetWorker V7.3

Disaster Recovery Guide

Chapter 15: HACMP for AIX Disaster Recovery

This chapter explains how to recover one or more computers in a High

Availability Cluster Multiprocessing for AIX (HACMP for AIX) environment.

This chapter includes the following sections:

"Recovering a Single Node" on page 191

"Recovering an Entire Cluster" on page 192

For detailed installation and configuration instructions, refer to the following documentation:

• EMC NetWorker Installation Guide, AIX Version

• NetWorker Administrator’s Guide

• HACMP for AIX documentation

Recovering a Single Node

This section describes how to recover a failed node within a cluster. This scenario assumes that the operating system on the participating nodes is intact and will not be recovered.

Recovering a cluster following a disaster is similar to any NetWorker software recovery. The important difference is that you must synchronize the data in the cluster database.

NetWorker V7.3

Disaster Recovery Guide

191

15

Chapter 15: HACMP for AIX Disaster Recovery

How to Recover a Single Node

To recover a single node:

1. Reinstall the NetWorker software.

2. Start the NetWorker client.

3. From an unaffected node in the cluster, use the virtual NetWorker server to recover the node's data and software from a recent backup.

4. From an unaffected node in the cluster, use the System Management

Interface Tool (SMIT) utility to synchronize the cluster topology and resources.

Recovering an Entire Cluster

This section describes how to perform a complete cluster recovery including reinstallation of the operating system on all nodes in the cluster.

Note:

It is unlikely that all the nodes in a cluster would fail simultaneously; but if this does occur, follow this procedure.

How to Perform a Cluster-Wide Recovery

Complete the following tasks to perform a cluster-wide recovery:

"Task 1: Reinstall the Software" on page 192

"Task 2: Recover the Cluster Database" on page 193

"Task 3: Start the Cluster" on page 193

"Task 4: Recover the Data" on page 194

Task 1: Reinstall the Software

To reinstall the software on a primary node:

1. Configure the hardware and reinstall the operating system.

2. Reinstall the HACMP for AIX software.

3. Reinstall the NetWorker software.

192

NetWorker V7.3

Disaster Recovery Guide

Chapter 15: HACMP for AIX Disaster Recovery

Task 2: Recover the Cluster Database

To recover the cluster database:

1. From all nodes in the cluster, make sure that all cluster software is stopped.

2. From one node in the cluster: a. Determine the save set ID that contains the following configuration directories:

– /etc/objres

– /usr/es/sbin/cluster/etc/objres

– /usr/es/sbin/cluster/etc/vg

Note:

The configuration directories in the previous list are valid for the

HACMP Extended Scalability (HACMP/ES) version of HACMP. The actual configuration directory path may vary depending on your version of HACMP.

If you do not know the save set ID, use the scanner command to determine its value, as follows:

scanner

device_path

> /tmp/savefile 2>&1

b. Enter the scanner -S commands to recover the cluster database, as follows:

scanner -S

saveset_id device_path

-x uasm -i Y -rv

/etc/objres scanner -S

saveset_id device_path

-x uasm -i Y -rv

/usr/es/sbin/cluster/etc/objres scanner -S

saveset_id device_path

-x uasm -i Y -rv

/usr/es/sbin/cluster/etc/vg

3. From the recovered node in the cluster, use the SMIT utility to synchronize the cluster topology and databases.

Task 3: Start the Cluster

To start the cluster:

1. Use the SMIT utility to bring up all nodes within the cluster.

2. Run the networker.cluster script on each node in the cluster.

3. Start the NetWorker resource group by using the SMIT utility.

15

NetWorker V7.3

Disaster Recovery Guide

193

15

Chapter 15: HACMP for AIX Disaster Recovery

Task 4: Recover the Data

To recover the data from the primary node:

1. If you do not know the save set ID of the most recent bootstrap, enter the

scanner -B

command to find it, as follows:

scanner -B

device_path

2. Enter the mmrecov command to recover the media database and resource files.

3. Complete the recovery of the resource database. In step 2, mmrecov recovered the resource directory to a temporary directory named res.R.

This is because the resource database cannot be overwritten while the

NetWorker software is running. To complete the recovery of the resource database: a. Stop the NetWorker resource group by using the SMIT utility.

b. Mount the shared disk containing the nsr directory.

c. Save a copy of the existing resource database, as follows:

mv /nsr/res /nsr/res.orig d. Place the recovered resource database files into the res directory, as follows:

mv /nsr/res.R /nsr/res

e. Unmount the shared disk containing the nsr directory.

f. Restart NetWorker resource group by using the SMIT utility.

4. Enter the nsrck -L7 command to recover the indexes.

5. Enter the recover command to recover the initialization directories.

Task 5: Create a Record of Backups Done after Bootstrap Created

After a disaster recovery, the media database will not contain information about any backups done after the bootstrap was created. The media database retains information about the last known file/record marks on the tape, and if backups were done after the bootstrap was created, the file/record information will not be current. For any tapes that contain backups that were completed after the bootstrap was created, use scanner with, at a minimum, the -m option to scan the save sets and/or tapes into the media database. This will ensure that the media database has a record of all completed save sets and that valid backups will not be overwritten by future backups.

194

NetWorker V7.3

Disaster Recovery Guide

Chapter 15: HACMP for AIX Disaster Recovery

Additionally, write protect all appendable volumes until you can complete the scanner activity to prevent scheduled backups from using these volumes. See

"Task 8: Create a Record of Backups Done after Bootstrap Created" on page 37

for instructions on write protecting and scanning.

15

NetWorker V7.3

Disaster Recovery Guide

195

15

Chapter 15: HACMP for AIX Disaster Recovery

196

NetWorker V7.3

Disaster Recovery Guide

Glossary

This glossary provides definitions for terms used in this guide.

Access Control List

(ACL)

ACL active group ad hoc backup administrator

Administrators group agent annotation

A list that specifies the permissions assigned to a specific file or directory.

See

Access Control List (ACL)

.

A NetWorker backup group that has its Autostart attribute enabled.

A backup that a user requests from the client. The user specifies the files, file systems, and directories to backup.

The person who installs, configures, and maintains the NetWorker software.

A Microsoft Windows user group whose members have the rights and privileges of users in other groups, plus the ability to create and manage the users and groups in the domain.

A term used by Sun Microsystems to denote a cluster server. Also known as a logical server (HP

TruCluster), a package (HP-UX), and a virtual server (Microsoft).

1. A comment associated with an archive save set.

2. A comment associated with an event.

NetWorker V7.3

Disaster Recovery Guide

197

Glossary

198

Application Specific

Module (ASM) archive archive request

A program that is used in a directive to specify how a set of files or directories is to be backed up or recovered. For example, compressasm is a

NetWorker directive used to compress files.

Backing up directories or files to an archive volume to free disk space. Archived data is not

recyclable. See also

groom

.

A NetWorker resource used to schedule and manage archiving.

archive volume

ASM attribute

A volume used to store archive data. You cannot store archive data on a

backup volume

or a

clone volume

.

See

Application Specific Module (ASM)

.

A feature (property) of a

resource

.

authorization code

A unique code that unlocks the NetWorker software for permanent use.

autochanger

See

library

.

autochanger sharing

See

library sharing

.

auto media management

A feature that enables the storage device to automatically label, mount, and overwrite an unlabeled or recyclable volume.

backup backup cycle backup group backup level

Backup Operators group

An operation that saves data to a volume.

The period of time from one level full backup to the next level full backup.

See

group

.

See

level

.

A Microsoft Windows user group that can log in to a domain from a computer or a server, and back up and restore its data. Backup operators can also shut down servers or computers.

NetWorker V7.3

Disaster Recovery Guide

Glossary

backup volume base enabler code bootstrap browse policy canned report carousel client client file index client-initiated backup

Client resource

A volume used to store backup data. You cannot store backup data on an

archive volume

or a

clone volume

.

See

enabler code

.

A save set that is essential for NetWorker disaster recovery. A bootstrap is composed of two components that reside on the NetWorker server:

the

media database

, and the

resource database

. The

bootstrap also provides information that is essential for recovering

client file indexes

.

A policy that determines how long backed up data is available for quick access. See also

retention policy

.

A preconfigured report that can be tailored by the user.

See

library

.

A computer, workstation, or fileserver whose data can be backed up or recovered. See also

Client resource

.

A database that tracks every database object, file, or

file system

that is backed up. The NetWorker server maintains a single client index file for each client.

See

ad hoc backup

.

Identifies the save sets to be backed up on a client.

The Client resource also specifies information about the backup, such as the schedule, browse policy, and retention policy for the save sets.

Multiple Client resources can be configured for one client computer. For example, you could create one Client resource to back up business data and another to back up operating system

files. See also

client

.

NetWorker V7.3

Disaster Recovery Guide

199

Glossary

200

clone clone volume cluster command-prompt

Common Internet

File System (CIF) connection port

Console server consolidate continued save set control zone daemon

DAR

A reliable copy of backed up data. Unlike volumes created with a simple copy command, clone volumes can be used in exactly the same way as the original backup volume. Single save sets or entire volumes can be cloned.

A volume used to store clone data. You cannot store clone data on an

archive volume

or a

backup volume

.

Two or more nodes that are connected and appear to network users as a single, highly available system.

The shell prompt, where you enter commands.

Formerly known as Server Message Block (SMB).

A message format used by DOS and Windows to share files, directories, and devices.

A port that NetWorker processes use to perform backup and recovery sessions through a firewall.

NetWorker servers and clients are managed from the NetWorker Console server. The Console server also provides reporting and monitoring capabilities for all NetWorker servers and clients.

To create a full backup by merging a new level 1 backup with the last full level backup.

Save set data that is continued from a previous volume.

A group of datazones managed by the NetWorker software.

A process that runs in the background and performs a specified operation at predefined times or in response to certain events.

See

Direct Access Restore (DAR)

.

NetWorker V7.3

Disaster Recovery Guide

Glossary

Data Management

Application (DMA)

Initiates a backup or recovery request through the

NDMP connection.

data retention policy

See

retention policy

.

Data Server Agent

(DSA)

An agent save or agent recover between a

NetWorker server and a non-NetWorker client. An example of a DSA is an NDMP host that generates proprietary save data and sends that data to a

NetWorker storage device to have a save set associated to it.

Data Service

Provider (DSP) datawheel datazone

DDS device

Controls access to disk storage during an NDMP back up.

See

library

.

A group of computers administered by a

NetWorker server.

See

Dynamic Drive Sharing (DDS)

.

1. A storage unit that reads from and writes to backup volumes. A storage unit can be a tape

device, optical drive,

library

, or file connected to

the

server

or

storage node

.

2. When dynamic drive sharing (DDS) is enabled, refers to the access path to the physical drive.

Device Central

DFS

DFS component

Direct Access

Restore (DAR)

The interface from which one can manage all

NetWorker libraries.

See

Distributed File System (DFS)

.

1. A namespace for files and DFS links, called a

DFS root.

2. A connection to a shared file or folder, called a

DFS child node

An NDMP data recovery operation that allows you to recover data in the middle of a tape set without having to parse the tape set sequentially.

This reduces recovery time of large backups.

NetWorker V7.3

Disaster Recovery Guide

201

Glossary

202

directed recovery directive

Distributed File

System (DFS)

DMA document mode domain controller drill-down reports drive

DSA

DSP

DSA save set

Dynamic Drive

Sharing (DDS)

EMC User SCSI

(LUS)

A recovery method used to recover data that originated on one computer to another computer.

Instructions to take special actions on a given set of files for a specified client during a backup.

A Microsoft Windows add-on that allows you to create a logical directory of shared directories that span multiple machines across a network.

See

Data Management Application (DMA)

.

A mode that displays static reports (as charts or tables) in a format that resembles the Print

Preview view displayed by a PDF file viewer.

A computer that manages user interactions with a domain, including logon, authentication, directory searches, and access to other shared resources.

Organizes basic report data in terms of granularity. For example, a user could start with group summary data, then "drill down" to data for a client within the group, and then to data for a selected save set.

When dynamic drive sharing (DDS) is enabled, refers to the physical backup object, such as a tape drive, disk, or file. See also

device

.

See

Data Server Agent (DSA)

.

Save sets of an NDMP client that are backed up to non-NDMP tape device.

See

Data Service Provider (DSP)

.

A feature that allows EMC NetWorker software to recognize shared drives.

A proprietary device driver that sends arbitrary

SCSI commands to a library.

NetWorker V7.3

Disaster Recovery Guide

enabler code enterprise event

Exit code expiration date expired save set failover file index file system firewall folder

NetWorker V7.3

Disaster Recovery Guide

Glossary

A special code that activates the software. The enabler code that unlocks the base features for software you purchase is referred to as a base enabler. Enabler codes for additional features or products (for example, library support) are

referred to as add-on enablers. See also

license string

.

Computers and folders organized into a tree-based visual representation.

A notification of an application event. This notification contains information that may require user action.

An indicator that specifies whether a backup or recovery session succeeded. An exit code of zero

(0) indicates the session completed successfully.

An exit code other than zero indicates that the session did not complete successfully.

The date when a volume changes from read/write to read-only.

A save set that has reached its browse time and is no longer browsable.

Relocating a cluster resource to its redundant

(backup) component, either because of a hardware or software failure or for administrative purposes.

See

client file index

.

1. A file tree located on a specific disk partition or other mount point.

2. The entire set of all files.

3. A method of storing files.

A system designed to prevent unauthorized access to or from a private network.

The graphical representation of a branching node in the Enterprise.

203

Glossary

204

full backup

Generic Services

Toolkit (GST) groom group

GST heterogeneous networks highly available system high-water mark host ID inactivity timeout incremental individual user authentication insertion time

Interactive mode

JAR

See

level

.

Provides the underlying framework of the

Console server.

To removing the original files from a disk after an archive.

A client or group of clients configured to back up files to the NetWorker server at a designated time of day.

See

Generic Services Toolkit (GST)

.

Computers of different platforms that interact across the network.

A system that allows cluster-aware services to continue despite a hardware or software failure.

The percentage of disk space that when filled, automatically starts the

staging

process.

A serial number that uniquely identifies a computer.

The number of minutes to wait before a client is considered to be unavailable for backup.

See

level

.

The process by which Console administrators restrict or grant user access to NetWorker servers, based on Console usernames.

The time that the save set record was most recently introduced into the save set database.

A mode that displays reports (as charts or tables) that users can interact with. For example, one can sort, rearrange, and resize columns in a table-format report that was run in this mode.

See

Java Archive (JAR)

.

NetWorker V7.3

Disaster Recovery Guide

Glossary

Java

Java Archive (JAR)

Java Plug-in

A high-level programming language. The

Java

Virtual Machine (JVM)

enables the same,

unmodified, Java program to run on most computer operating systems.

A file that contains compressed components needed for a Java applet or application.

A JVM that can be used by a web browser to run

Java applets.

Java Virtual Machine

(JVM)

An execution environment for interpreting the

Java programming language. Each operating system runs a unique JVM to interpret Java code.

jukebox

JVM

See

library

.

See

Java Virtual Machine (JVM)

.

label legacy method

A NetWorker assigned label that uniquely identifies a volume. Templates can be used to define label parameters.

The use of special-case Microsoft APIs to back up and recover operating system components, services, and applications.

level library

A measurement that determines how much data is saved during a scheduled or ad hoc backup.

A full (f) backup backs up all files, regardless of whether they have changed. Levels one through nine [1-9] back up files that have changed since the last lower numbered backup level. An incremental (incr) backup backs up only files that have changed since the last backup.

A mechanism that uses a robotic arm to move media among various components in a device.

The term library is synonymous with autochanger, autoloader, carousel, datawheel, jukebox, and near-line storage.

NetWorker V7.3

Disaster Recovery Guide

205

Glossary

206

library sharing license file license string

License Manager local cluster client locale settings

Shared access of servers and storage nodes to the tape drives in a library.

A license file contains one or more NetWorker

license string

s. Licenses ordered from FSC are provided to the customer via NELI (NetWorker

Licensing tool). Customers download their license file from the web.

Describes a NetWorker order unit. It contain a unique key and checksum. The keylic command generates the appropriate license resource with an enabler code from a license string.

An application that provides a central point for managing product licenses.

A NetWorker client that is not bound to a physical machine, but is instead managed by a cluster manager. It is also referred to as a logical or virtual client.

Settings that specify the input and output formats for date and time, based on local language conventions.

logical cluster client

See

virtual cluster client

.

logical device

A specific NetWorker device used in the integration of NetWorker software with

AlphaStor. AlphaStor assigns the logical device to an actual device.

logical server

A term used by HP TruCluster to denote a cluster server. Also known as an agent (Sun), a package

(HP-UX), and a virtual server (Microsoft).

low-water mark

The percentage of disk space filled that, when reached, automatically stops the staging process.

See

EMC User SCSI (LUS)

.

LUS managed application

A program that can be monitored and/or administered from the Console server.

NetWorker V7.3

Disaster Recovery Guide

Glossary

manual backup media media database media index member metadata mount mount point multiplexing

NDMP

NDMP data server

NDMP host

NDMP save set

See

ad hoc backup

.

The physical storage medium, such as magnetic tape, optical disk, or file system to which backup data is written.

Indexed entries about the location and the life cycle status of all data and volumes that the

NetWorker server manages. See also

volume

.

See

media database

.

A physical host, such as the HP TruCluster Server, in a cluster environment. Each member has its own IP address. A member can also be a

virtual server

.

VSS-defined information that is passed from the

writer

to the

requestor

. Metadata includes the

writer name, a list of

VSS components

to back up, a list of components to exclude from the backup, and the methods to use for recovery.

To make a volume physically available so that backup data can be written to it.

See

volume mount point

.

To simultaneously write data from more than one save set to the same storage device.

See

Network Data Management Protocol (NDMP)

.

An NDMP service that either reads from or writes to disk.

The host computer that executes the NDMP server application.

A save set of an NDMP client that is backed up to an NDMP tape device.

NetWorker V7.3

Disaster Recovery Guide

207

Glossary

208

NDMP server

NDMP service

An instance of one or more NDMP services controlled by a single NDMP control connection.

Thus a data/tape/SCSI server is an NDMP server providing data, tape, and SCSI services.

The virtual state machine on an NDMP host that is controlled by the DMA. Examples of NDMP services include:

• A general server with direct attached storage

(storage appliance)

• A system with one or more tape drives

• A software process that reads two data streams and multiplexes them into one stream

NDMP storage node

A host having NDMP services or any open system having NDMP services installed. For instance,

Netapp Filer and EMC Filer.

near-line storage

See

library

.

Network Data

Management

Protocol (NDMP)

Network File System

(NFS)

A TCP/IP-based protocol that specifies how heterogeneous network components communicate for the purposes of backup and recovery.

A client/server application that allows users to access shared files stored on networked computers of different types.

NetWorker administrator

NetWorker client

NetWorker Console server

NetWorker

Management

Console

A user who can add to or change the configuration of the NetWorker server, media devices, and libraries. NetWorker administrators must have their usernames included in the

NetWorker server Administrator list.

See

client

.

See

Console server

.

See

Console server

.

NetWorker V7.3

Disaster Recovery Guide

Glossary

NetWorker server

NetWorker storage node

NFS

NFS server

The computer running the NetWorker server

software, which contains the

online indexes

and

provides backup and recovery services to the clients on the same network.

See

storage node

.

node nonclone pool non-NDMP device notification nsrd nsrhost online indexes operator override package packet filtering

See

Network File System (NFS)

.

A computer that contains exported file systems that NFS clients can access.

See

member

.

Pools that contain data that has not been cloned.

A device that is not configured as an NDMP type of device.

A message sent to the NetWorker administrator about important NetWorker events.

The master NetWorker server process.

The logical hostname of the NetWorker server.

Databases on the NetWorker server that contain information about client backups (

client file index

)

and backup volumes (

media database

).

The person who monitors the server status, loads backup volumes into storage devices, and executes day-to-day NetWorker tasks.

A backup level that takes place instead of the scheduled one.

A term used by HP-UX to denote a cluster server.

Also known as an agent (Sun), logical server (HP

TruCluster), and virtual server (Microsoft).

A method of firewall protection that accepts or rejects packets entering or leaving the network

based on user-defined rules. See also

firewall

.

NetWorker V7.3

Disaster Recovery Guide

209

Glossary

210

parallelism pathname peer physical cluster client physical host policy pool

PowerSnap probe purging recover recyclable save set

A method of backing up or recovering data from several clients or save sets in parallel.

Instructions for accessing a file. An absolute

pathname indicates how to find a file starting from the root directory. A relative pathname indicates how to find the file starting from the current directory.

A NetWorker host that is involved in an authentication process with another NetWorker host.

A NetWorker client that is bound to a physical host in the cluster and can have its own resources

(private or local).

A node or host that forms part of a cluster.

A set of constraints that specify how long data is available for recovery. Each Client resource has a browse policy and a retention policy.

A feature to sort backup data to selected volumes.

Technology that provides point-in-time snapshots of data. NetWorker software backs up data from the snapshot. This allows applications to continue to write data during the backup operation, and ensures that open files are not omitted.

A query to determine the directories or files to back up on each client.

Deleting file entries from the client file index.

To restore files from a backup volume to a client disk.

A save set whose browse and retention policies have expired. Recyclable save sets are removed from the media database.

NetWorker V7.3

Disaster Recovery Guide

Glossary

recyclable volume

Registry remote device

Remote Procedure

Call (RPC) requestor resource resource database resource owner retention policy retrieve retry mechanism root

A volume whose data has passed both its browse and retention policies and is now available to be relabeled.

A Microsoft Windows database that centralizes all

Windows settings and provides security and control over system, security, and user account settings.

A storage device that is attached to a storage node.

The protocol that the NetWorker server uses to perform client requests over a network.

A VSS-aware application that creates and destroys a

shadow copy

. NetWorker software is a requestor.

A component that describes the NetWorker server or its clients. Clients, devices, schedules, groups, and policies are all NetWorker resources. Each resource has attributes that define its properties.

A database of information about each resource.

A logical cluster host that owns the resource. If a

Cluster resource, such as a shared disk, is not owned by a virtual host, it is assumed to be owned by the physical node that hosts the resource.

Determines how long backup data is available for recovery, though not necessarily immediate

recovery. See also

browse policy

.

To locate and recover archived files and directories.

The action NetWorker software performs when client operations fail. This situation might occur when the rate of transmission is either low or nonexistent.

1. (UNIX only) The UNIX superuser account.

2. (Microsoft Windows and UNIX) The highest level of the system directory structure.

NetWorker V7.3

Disaster Recovery Guide

211

Glossary

212

RPC

SAN Reservation

Service (SRS) save save set save set consolidation save set ID save set recover save set status save stream scanner server server index service port shadow copy

See

Remote Procedure Call (RPC)

.

SRS is a Fujitsu Siemens Computers product that allows backup devices in a SAN to be shared between storage nodes.

The command that backs up client files and makes entries in the online index.

A group of files or a file system that is backed up on storage media.

A process that performs a level 1 backup and merges it with the last full backup of a save set to create a new full backup. See also

level

.

An internal identification number assigned to a save set.

To recover data by specifying save sets rather than by browsing and selecting files or directories.

An attribute that indicates whether a save set is browsable, recoverable, or recyclable. The save set status also indicates whether the save set was successfully backed up.

The data and save set information being written to a storage volume during a backup.

A command used to read a backup volume when

the

online indexes

are no longer available.

See

NetWorker server

and

Console server

.

See

client file index

.

A port used to listen for backup and recover requests from clients through a firewall.

A temporary, point-in-time copy of a volume created using VSS technology. See also

Volume

Shadow Copy Service (VSS)

.

NetWorker V7.3

Disaster Recovery Guide

Glossary

shared disk shell prompt silo

Simple Network

Management

Protocol (SNMP) skip snapshot snapshot policy

SNMP

SRS ssid staging stand-alone stand-alone device storage device

The storage disk that is shared among multiple nodes in a cluster.

A cue in a shell window where you enter a command.

A repository for holding hundreds or thousands of volumes. Silo volumes are identified by bar codes, not by slot numbers.

A protocol used to send messages to the administrator about NetWorker events.

A backup

level

in which designated files are not

backed up.

A point-in-time copy of data created during an instant backup.

A NetWorker resource used to configure backups using the NetWorker PowerSnap Module software.

See

Simple Network Management Protocol (SNMP)

.

See

SAN Reservation Service (SRS)

.

See

save set ID

.

Moving data from one storage medium to a less-costly medium, and later removing the data from its original location.

In a cluster environment, if the

NetWorker.clustersvr file is missing, the NetWorker server starts in noncluster (stand-alone) mode.

A storage device that contains a single drive for backing up data.

The hardware that reads and writes data during backup, recovery, or other NetWorker operations.

NetWorker V7.3

Disaster Recovery Guide

213

Glossary

storage node tape service user alias user groups versions target sessions

TCP/IP

Transmission

Control

Protocol/Internet

Protocol (TCP/IP) trap user

A storage device physically attached to another computer whose backup operations are controlled by the NetWorker server.

An NDMP DSP service that controls access to tape storage. A system can simultaneously host multiple tape services corresponding to multiple backup streams.

The number of backup sessions accepted by a backup device.

See

Transmission Control Protocol/Internet Protocol

(TCP/IP)

.

A standard set of communication protocols that connects hosts on the Internet.

Unsolicited notification sent from the SNMP agent to the network manager’s SNMP event manager.

1. A NetWorker user who can back up and recover files from a computer.

2. A Console user who has standard access privileges to the Console server.

The username seen by the NetWorker server when a Console user connects to the NetWorker server.

Used to assign users to access control groups and to configure the privileges associated with those groups.

The date-stamped collection of available backups for any single file.

214

NetWorker V7.3

Disaster Recovery Guide

Glossary

virtual cluster client

A NetWorker client that is not permanently bound to a physical machine, but is instead managed by a cluster manager. It can be bound to more than one physical machine in the cluster and can own its own data disks. It is also referred to as a logical cluster client or a virtual client.

virtual server

A Microsoft term to denote a cluster server. Also known as an agent (Sun), a logical server (HP

TruCluster), and a package (HP-UX).

volume

1. The physical storage medium, such as magnetic tape, optical disk, or file system to which backup data is written.

2. An identifiable unit of data storage that may reside on one or more computer disks.

volume ID volume mount point

A disk volume that is grafted into the namespace of a host disk volume. This allows multiple disk volumes to be linked into a single directory tree, and a single disk or partition to be linked to more than one directory tree.

volume name

An internal identification that NetWorker software assigns to a backup volume.

The name assigned to a backup volume when it is

labeled. See also

label

.

See

pool

.

volume pool

Volume Shadow

Copy Service (VSS)

VSS

VSS component

Microsoft technology that creates a point-in-time snapshot of a disk volume. NetWorker software backs up data from the snapshot. This allows applications to continue to write data during the backup operation, and ensures that open files are not omitted

See

Volume Shadow Copy Service (VSS)

.

A subordinate unit of a

writer

.

NetWorker V7.3

Disaster Recovery Guide

215

Glossary

writer

A database, system service, or application code that works with VSS to provide

metadata

about

what to back up and how to handle

VSS components

and applications during backup and restore.

216

NetWorker V7.3

Disaster Recovery Guide

Index

A

AIX HACMP cluster, recovering

AIX, recovering FullTime AutoStart

191

179

application data

recovering on NetWorker client

UNIX

Windows

43

, 98

89

recovering on NetWorker server

NetWare

111

UNIX

Windows

36

83

recovering on storage node

UNIX

40

Windows

autochanger

89

NetWare, resource, configuring

UNIX resetting

Windows

31

108

removable storage disabled, restoring with

74

removable storage enabled, restoring with

85

resetting

79

automated system recovery (ASR) on

Windows

55

B backup

importance of scheduling testing

UNIX

Windows

38

, 42

, 45

, 100

86 ,

92

17

boot -file genvmunix, HP TruCluster bootstrap

locating save set ID of

NetWare

109

UNIX

Windows

31

75

162

NetWorker V7.3

Disaster Recovery Guide recovering

NetWare

UNIX

Windows

32

109

77

browse policy, Windows

C

75

chkdsk, (Windows 2000 or 2003)

145

client

See NetWorker client

client file indexes

recovering

NetWare

UNIX

35

109

Windows

81

clu_add_member, HP TruCluster

162

clu_create, HP TruCluster

170

clu_delete_member -m, HP TruCluster

162

cluster, (Windows 2000 or 2003)

cluster database definition

139

prerequisites for recovery recovering a single node

139

141

recovering failed quorum disk recovering multiple nodes

144

140

cluster, FullTime AutoStart for AIX, HP-UX,

Linux, and Solaris

recovering a single node recovering all nodes

181

179

renaming resource directory

186

cluster, FullTime AutoStart for Windows

recovering a single nodes

cluster, HACMP for AIX

189

recovering a single node recovering all nodes

192

191

cluster, HP TruCluster

prerequisites for recovery recovering a single node adding node

161

159 –

160

160

hardware mappings, verifying partitioning boot drive

161

recovering unshared node date

161

162

217

Index

C - C removing/re-adding node replacing hardware

161

recovering all nodes

163

configure cluster software

161

164

configure NetWorker software hardware mappings, verifying install NetWorker software locating bootstrap

166

165

166

164

locating client file index recovering bootstrap

167

169

recovering client file index recovering the cluster data

169

170

re-creating the cluster replacing hardware

170

164

cluster, HP-UX

recovering all nodes recovering single node

cluster, Sun

176

175

prerequisites

155

recovering all nodes recovering single node

cluster, VERITAS

156

155

prerequisites

149

recovering all nodes

151

recovering single node

150

cluster, Windows NT 4.0 Cluster

recovering all nodes

131

recovering MSCS configuration recovering quorum disk recovering shared disk

128

125

recovering single node

123

cmapplycomf, HP-UX cluster

177

cmdeleteconf, HP-UX cluster cmgetconf, HP-UX cluster

177

177

cmhaltpkg, HP-UX cluster cmmodpkg, HP-UX cluster

178

178

cmruncl, HP-UX cluster cmrunnode, HP-UX cluster

177

175

129

commands

FullTime AutoStart for AIX, HP-UX, Linux, and

Solaris envsh

183 ,

185

scanner

182

scanner -S -x uasm -rv -i Y

HACMP for AIX mmrecov

194

nrsck -L7

194

183

218 recover scanner -B

194

194

scanner -S -x uasm -i Y -rv

HP TruCluster boot -file genvmunix clu_add_member

162

162

clu_create

170

clu_delete_member -m dfsmgr -e

165

hwmgr -view

160 ,

164

162

193

inquire

167

mminfo -av -N

169

mminfo -mv nsrck -L6

170

166

nsrim -X nsrinfo

170

170

nsrjb -I nsrjb -vHE

166

166

recover -c recover -S -iY

162 ,

171 –

172

169

rmfdmn scanner -B sjiielm

HP-UX

162

167

166

cmapplycomf cmdeleteconf cmgetconf cmhalpkg

177

178

177

177

cmmodpkg cmruncl

177

178

cmrunnode exchange

175

178

mount newfs scanner -B umount

178

177

176

178

vgchange

LINUX, ielem

177 –

178

31

Mac OS X, recover

NetWare

98

dsrepair install

102

,

114 –

115 ,

117 ,

119 ,

121

102 ,

104 ,

116 ,

119

load NETUTIL

LOAD TSA

109

116

– 117 ,

121

ndir

102 ,

114 ,

117 ,

121

netadmin

102 ,

115

nwadmin

102

, 114 –

115

, 117

NetWorker V7.3

Disaster Recovery Guide

nwconfig nwdown

102

, 104

,

116 ,

119

111

partmgr

115

rights /T /S scanner -B

UNIX

121

109

inquire

32

mminfo -avot

40

,

43 ,

98

mminfo -mv mmrecov

31

32 ,

34

mt

27

nsr_shutdown -a

30 ,

34

nsradmin nsrck

37

,

42 ,

45

32 ,

36

nsrck -L7 nsrd

35

30

,

35

nsrexecd nsrjb -HE

30 ,

35

35

nsrjb -l

31

nsrjb -lnv -S -f

32

nsrjb -lv nsrjb -vHE nsrmm -d nwrecover ping

27

scanner -B sjiielm

Windows

31

35

31

35

36

,

40

, 43

32

inquire

76 –

77

mminfo -mv

57 ,

71

mminfo -s -mv mmrecov

88

51 ,

53 ,

73 ,

75 ,

77 ,

79 ,

86

nslookup nsradmin nsrck nsrck -L7

77

64 ,

70

85

,

92

52 ,

82

nsrjb

80

nsrjb -lnv -S nsrjb -lnv -S -f nsrjb -lv

52

nsrjb -lv -S nsrjb -vHE

51

76 –

77

51 ,

74 ,

80

51 –

52 ,

80

nsrmm -d

78 ,

82

nsrmmdbasm

79

ping

64

,

70

recover -c

63

recover.exe

62

NetWorker V7.3

Disaster Recovery Guide scanner scanner -B sjiielm uasm

80

79

76

51 ,

76

Windows 2000 or 2003 cluster chkdsk

145

net start clussvc net stop clussvc

144

143 ,

145 –

146

Windows NT 4.0

-fixquorum

136

regedit

135

computer data, recovering

See recovering

configuration files

recovering for NetWorker server

NetWare

111

UNIX

Windows

34

– 35

79

configuring NetWorker server

NetWare

108

UNIX

Windows

30

73

configuring operating system

NetWare

105

UNIX

Windows

24

26

69

Console software, recovering corrupted database

46

,

95

customer

feedback service

14

14

46 ,

95

D daemons, shutting down NetWorker data, recovering

See recovering

database

corrupted

46

,

95

recovering

46 ,

95

database, resource and media

bootstrap for

NetWare

109

UNIX

Windows

31

– 32

75 ,

77

device

NetWare, configuring

108

30

Index

D - D

219

Index

E - I

UNIX configuring testing

Windows

27

30

configuring testing

69

69 ,

79

dfsmgr -e, HP TruCluster disaster

165

definition preparing for

16

17

recovery prerequisites

disaster recovery

46 ,

95

documentation

12

drive

18

replacing, NetWare replacing, UNIX

23

101

replacing, Windows

drives, phantom

135

55

dsrepair, NetWare

102

, 114

– 115 ,

117 ,

119 ,

121

Dynamic Host Configuration Protocol (DHCP) database

93

E electronic access to Fujitsu Siemens information

14

Emergency repair disk

creating, Windows NT using, Windows NT

end user data

See user data

63

61

envsh, FullTime AutoStart for AIX, HP-UX,

Linux, and Solaris

183 ,

185

exchangec, HP-UX cluster

178

F feedback

14

file, legatomm firewall support

113

203

-fixquorum, Windows NT 4.0

Fujitsu Siemens

customer feedback customer service

14

14

support

14

FullTime AutoStart Cluster

136

AIX, HP-UX, Linux, and Solaris recovering all nodes

181

recovering single node

Windows

179

recovering a single node

189

FullTime AutoStart for AIX, HP-UX, Linux, and

Solaris

recovering a single node recovering all nodes

181

179

renaming the resource directory

186

FullTime AutoStart for Windows, recovering single node

189

G gathering key information before a disaster

H

18

HACMP for AIX cluster

restoring a single node restoring all nodes

192

191

hard drive

101 –

102

replacing, NetWare replacing, UNIX replacing, Windows

23

55

HP TruCluster

recovering a single node

160

recovery prerequisites

159

See also cluster, HP TruCluster

HP-UX Cluster

recovering all nodes

176

recovering single node

175

HP-UX, recovering FullTime AutoStart hwmgr -view, HP TruCluster

160

,

164

179

I ielem, Linux

31

information before a disaster, gathering key

18

information, recovering

See recovering

inquire, HP TruCluster inquire, UNIX

32

inquire, Windows

167

76

77

220

NetWorker V7.3

Disaster Recovery Guide

install command, NetWare installing

102 ,

104

NetWorker client

UNIX

43

, 98

Windows

89

NetWorker server

NetWare

UNIX

29

108

Windows

72

NetWorker storage node

UNIX

Windows

39

89

operating system

UNIX

24

– 26

, 104

Windows

66

IntraNetWare SYS Volume, recovering introduction

15

inventory autochanger

UNIX

Windows

31

79

,

116

,

119

118

J jukebox

See autochanger

K key information before a disaster, gathering

18

L legatomm file

113

limitations, Windows OEM recovery CD

Linux, recovering FullTime AutoStart

179

56

load NETUTIL, NetWare

109

LOAD TSA, NetWare

116 –

117 ,

121

locating, server’s bootstrap save set ID

109

NetWare

UNIX

31

Windows

75

Index

J - N

M media database

locating bootstrap for

NetWare

UNIX

31

109

Windows

75

Microsoft automated system recovery

(ASR)

55

Microsoft cluster configuration, recovering,

Windows NT 4.0

129

Microsoft Cluster Server, recovering

Windows 2000 or 2003

Windows NT 4.0

123

139

mminfo -av -N, HP TruCluster mminfo -avot, UNIX

40

,

43 ,

98

mminfo -mv, HP TruCluster mminfo -mv, UNIX

31

169

166

mminfo -mv, Windows mminfo -s -mv, Windows mmrecov, HACMP for AIX mmrecov, UNIX

32

, 34

57 ,

71

88

194

mmrecov, Windows mount, HP-UX cluster

51 ,

53 ,

73 ,

75 ,

77 ,

79 ,

86

178

mt, UNIX

27

multiple disaster recoveries

performing in UNIX performing in Windows

22

54

N ndir, NetWare

102

, 114 ,

117

,

121

NDMP (Network Data Management Protocol)

DSA

DSP

201

201

NDMP data server

NDMP host

207

NDMP save set

NDMP server

207

208

207

NDMP service

208

NDMP storage node non-NDMP device

NDS data, recovering

208

209

112

NDS partition, recovering replicated

114

net start clussvc, (Windows 2000 or 2003)

144

NetWorker V7.3

Disaster Recovery Guide

221

Index

N - N

net stop clussvc, (Windows 2000 or 2003)

145 –

146

netadmin, NetWare

102

, 115

NetWare

legatomm file

113

locating bootstrap

109

143

NetWorker server recovery prerequisites

107

prerequisites summary recovering

20

bootstrap

109

IntraNetWare SYS volume

NetWare 4.10 SYS volume

NetWare 4.11 SYS volume

NetWare 5.0 SYS volume

NetWorker server

107

operating system

103

operating system prerequisites replicated NDS partition recovering NDS data

112

114

118

115

118

118

103

renaming the NetWorker server configuration files

111

replacing a hard drive

NetWorker client

101

, installing

UNIX

Windows recovering

43

, 98

89

UNIX

Windows

42

87

NetWorker server

configuring

NetWare

UNIX

30

NetWare

UNIX

28

108

Windows installing

Windows recovering

73

NetWare

UNIX

29

NetWare

UNIX

31

108

Windows

72

locating bootstrap save set ID

109

75

107

Windows

71

recovering configuration files

NetWare

111

UNIX

Windows

34

– 35

79

recovering user data on

UNIX

36

Windows

83

testing backup and recovery

NetWare

UNIX

38

113

Windows

86

NetWorker storage node

installing

UNIX

39

Windows recovering

UNIX

Windows

38

89

87

NetWorker Utilities program, NetWare newfs, HP-UX cluster

177

nslookup, Windows nsr_shutdown -a, UNIX

64 ,

70

30 ,

34

nsradmin, UNIX

37 ,

42 ,

45 ,

85 ,

92

nsrck

-L6

HP TruCluster

170

-L7

HACMP for AIX

UNIX

Windows

35

UNIX

Windows

32

,

36

77

52 ,

82

194

nsrd, UNIX

30 ,

35

nsrexecd, UNIX

30

, 35

nsrim -X, HP TruCluster nsrinfo, HP TruCluster

170

170

nsrjb

-HE option

UNIX

-I option

35

HP TruCluster

-Inv -S options

Windows

-Iv option

51

Windows

-Iv -S options

52

Windows

166

51 ,

80

109

222

NetWorker V7.3

Disaster Recovery Guide

-l option

UNIX

31

-lnv -S -f options

UNIX

32

Windows

-lv option

76 –

77

UNIX

35

-lv -S option

Windows

-vHE option

74

51

HP TruCluster

UNIX

31

Windows

Windows

166

80

51 –

52

, 80

nsrmm -d

UNIX

35

Windows

78 ,

82

nsrmmdbasm, Windows

79

NT 4.0

repairing operating system with Emergency

Repair Disk

nwadmin, NetWare

63

102

, 114 –

115

,

117

nwconfig, NetWare nwdown, NetWare nwrecover, UNIX

102 ,

104 ,

116

,

119

111

36 ,

40

,

43 ,

98

O

OEM recovery CD limitations operating system

NetWare, recovering

103

UNIX installing recovering testing

27

24 –

26

24

Windows installing recovering testing

69

66

64

Windows NT 4.0, repairing

56

61

P partition, recovering a replicated NDS partition partmgr, NetWare

114

115

NetWorker V7.3

Disaster Recovery Guide

Index

O - Q

phantom drives physical host ping, UNIX ping, Windows

27

135

207

64 ,

70

policies

browse, Windows retention, Windows

preparing for a disaster

75

75

17

prerequisites

automated system recovery, Windows

57

hard drive, replacing

UNIX

Windows

23

55

56 –

HP TruCluster recovery

159

Microsoft cluster recovery, Windows 2000 or

2003

139

Microsoft configuration recovery, Windows

NT 4.0

129

NetWorker client recovery

Mac OS X

UNIX

42

97

Windows

88

NetWorker server recovery

NetWare

UNIX

28

107

Windows

71

NetWorker storage node recovery

UNIX

Windows

39

88

operating system recovery

NetWare

103

UNIX

Windows

24

65

Windows NT 4.0

summary of

18

Sun cluster recovery

61

155

VERITAS cluster recovery

149

program, NetWorker Utilities for

NetWare

109

Q quorum disk

recovering, Windows 2000 or 2003 recovering, Windows NT 4.0 Cluster

140

128

223

Index

R - R

R recover -c, HP TruCluster recover -c, Windows

63

162

recover -S -iY, HP TruCluster recover, HACMP for AIX

194

,

171

169

172

recover.exe, Windows consideration recovering

62

Console database

46 ,

95

46 ,

95

recovering bootstrap

UNIX

32

Windows

77

recovering cluster nodes

FullTime AutoStart cluster for Windows recovering a single node

189

FullTime AutoStart for AIX, HP-UX, Linux, and

Solaris all nodes single node

181

179

HACMP for AIX all nodes

192

single node

HP TruCluster

191

all nodes single node

163

160

HP-UX cluster all nodes

176

single node

Sun

175

all nodes single node

156

155

VERITAS all nodes

151

single node

150

Windows 2000 or 2003 multiple nodes quorum disk

144

140

141

singles node

Windows NT 4.0

all nodes

131

cluster configuration quorum disk shared disk

128

125

129

single node

123

recovering configuration files

NetWorker server

UNIX

Windows

34

– 35

79

recovering IntraNetWare SYS volume

118

recovering multiple computers, consideration

Windows

54

recovering NDS data

112

recovering NetWare 4.10 SYS volume recovering NetWare 4.11 SYS volume recovering NetWare 5.0 SYS volume recovering NetWorker client

115

118

118

UNIX

Windows

42

87

recovering NetWorker server

NetWare

107

UNIX

Windows

28

71

recovering NetWorker storage node

UNIX

38

Windows

87

recovering operating system

NetWare

UNIX

24

103

Windows

64

, 66

recovering volume mount points,

Windows

83 ,

89

recovering, replicated NDS partition recoverpsm recovery

46

, 95

114

summary steps

UNIX

22

Windows testing

50 ,

54

UNIX

Windows

38

, 42

, 45

, 100

60 ,

86 ,

92

recovery limitations, Windows OEM CD recovery prerequisites

18

recovery, Microsoft automated regedit, Windows NT 4.0

135

55

regular backups, importance of related documentation

12

17

Remote Access List, Windows NT

62

removable storage autochanger, restoring renaming NetWorker configuration files

NetWare

111

UNIX

Windows

34

35

79

56

85

224

NetWorker V7.3

Disaster Recovery Guide

Index

S - S

renaming the resource directory, FullTime

AutoStart for AIX, HP-UX, Linux, and

Solaris repair disk

186

creating

Windows NT using

Windows NT

61

63

repairing Windows NT 4.0 OS replacing hard drive

61

NetWare

UNIX

23

101

Windows

55

replacing the quorum disk

Windows NT 4.0 Cluster

128

replicated NDS partition, recovering

114

resetting autochanger

UNIX

31

Windows

79

resource database

locating bootstrap for

NetWare

109

UNIX

Windows

31

75

retention policy, Windows rights /T /S, NetWare

121

rmfdmn, HP TruCluster

75

162

S save set ID

locating for NetWorker server

NetWare

109

UNIX

Windows

31

75

scanner

-B option

HACMP for AIX

HP TruCluster

NetWare

UNIX

32

109

HP-UX cluster

-S -x uasm -i Y -rv

194

167

Windows

51 ,

76

FullTime AutoStart for AIX, HP-UX, Linux, and

Solaris

-S option

182

176

FullTime AutoStart for AIX, HP-UX, Linux, and Solaris

183

HACMP for AIX

Windows

76

193

scheduled backups, importance of

17

server

configuring NetWorker

NetWare

108

UNIX

Windows

30

73

installing NetWorker

NetWare

108

UNIX

Windows

29

72

recovering NetWorker

NetWare

107

UNIX

Windows

28

71

shutdown NetWorker, UNIX

service for customers

14

setting

browse policy, Windows retention policy, Windows

shutdown NetWorker server

UNIX

sjiielm

30

166

HP TruCluster

UNIX

31

Windows

storage node

80

75

75

30

See NetWorker storage node

summary recovery steps

UNIX

Windows

22

50

, 54

Sun cluster

restoring all nodes restoring single node

support

14

156

155

SYS volume, recovering

IntraNetWare

118

NetWare 4.10

NetWare 4.11

115

118

NetWare 5.0

118

system recovery, Microsoft automated system state, cluster database definition,

Windows 2000 or 2003

139

55

NetWorker V7.3

Disaster Recovery Guide

225

Index

T - W

T testing backup and recovery

NetWare

113

UNIX

Windows

38

,

42 ,

45

,

100

86

, 92

testing OS

NetWare

UNIX

Windows

27

106

69

TruCluster, recovering

159

See also cluster, HP TruCluster

U uasm, Windows

79

umount, HP-UX cluster

178

UNIX

configuring operating system installing operating system locating bootstrap

31

prerequisites summary recovering

19

24

26

26

bootstrap

32

NetWorker client

NetWorker server

42

28

NetWorker storage node

38

recovering operating system

26

renaming NetWorker server configuration files

34

summary of disaster recovery

22

user data

recovering on NetWorker client

Mac OS X

UNIX

43

98

Windows

89

recovering on NetWorker server

UNIX

Windows

36

83

recovering on storage node

UNIX

40

Windows

89

Utilities program, NetWare

109

226

V

VERITAS cluster

restoring all nodes restoring single node

vgchange, HP-UX cluster

151

150

177 –

178

volume mount points, recovering on

Windows

83

W web address

14

Windows

89

automated system recovery (ASR) installing operating system

66

55

locating bootstrap

75

NT 4.0, repairing OS with Emergency Repair

Disk

63

performing multiple disaster recoveries prerequisites summary recovering

19

a NetWorker server bootstrap

77

71

54

FullTime AutoStart Cluster for

Windows

189

NetWorker client

87

NetWorker storage node

87

operating system

66

volume mount points

83

,

89

renaming NetWorker server configuration files

79

replacing a hard drive setting

55

browse policy retention policy

75

75

summary of disaster recovery

50 ,

54

Windows Cluster, 2000 or 2003 recovery

multiple nodes prerequisites

144

139

140

quorum disk server

139

single node

141

Windows Cluster, NT 4.0

quorum disk recovering data replacing

128

128

NetWorker V7.3

Disaster Recovery Guide

recovery all nodes

131

cluster configuration one cluster node

123

129

shared disk

125

Windows Internet Naming Service (WINS) database

93

Windows NT repair disk

creating using

63

61

Index

W - W

NetWorker V7.3

Disaster Recovery Guide

227

Index

W - W

228

NetWorker V7.3

Disaster Recovery Guide

Fujitsu Siemens Computers GmbH

User Documentation

81730 Munich

Germany

Fax: (+49) 700 / 372 00000

e-mail: [email protected]

http://manuals.fujitsu-siemens.com

Submitted by

Comments on NetWorker V7.3, Disaster Recovery Guide

Comments

Suggestions

Corrections

NetWorker V7.3

Disaster Recovery Guide

Information on this document

On April 1, 2009, Fujitsu became the sole owner of Fujitsu Siemens Computers. This new subsidiary of Fujitsu has been renamed Fujitsu Technology Solutions.

This document from the document archive refers to a product version which was released a considerable time ago or which is no longer marketed.

Please note that all company references and copyrights in this document have been legally transferred to Fujitsu Technology Solutions.

Contact and support addresses will now be offered by Fujitsu Technology Solutions and have the format

…@ts.fujitsu.com

.

The Internet pages of Fujitsu Technology Solutions are available at

http://ts.fujitsu.com/

... and the user documentation at

http://manuals.ts.fujitsu.com

.

Copyright Fujitsu Technology Solutions, 2009

Hinweise zum vorliegenden Dokument

Zum 1. April 2009 ist Fujitsu Siemens Computers in den alleinigen Besitz von

Fujitsu übergegangen. Diese neue Tochtergesellschaft von Fujitsu trägt seitdem den Namen Fujitsu Technology Solutions.

Das vorliegende Dokument aus dem Dokumentenarchiv bezieht sich auf eine bereits vor längerer Zeit freigegebene oder nicht mehr im Vertrieb befindliche

Produktversion.

Bitte beachten Sie, dass alle Firmenbezüge und Copyrights im vorliegenden

Dokument rechtlich auf Fujitsu Technology Solutions übergegangen sind.

Kontakt- und Supportadressen werden nun von Fujitsu Technology Solutions angeboten und haben die Form

…@ts.fujitsu.com

.

Die Internetseiten von Fujitsu Technology Solutions finden Sie unter

http://de.ts.fujitsu.com/

..., und unter

http://manuals.ts.fujitsu.com

finden Sie die

Benutzerdokumentation.

Copyright Fujitsu Technology Solutions, 2009

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement

Table of contents