Avamar EMC. Data Migration Enabler. User Guide. Version 7.4 and Service Packs REV 02

Transcription

1 EMC Avamar Version 7.4 and Service Packs Data Migration Enabler User Guide REV 02

2 Copyright EMC Corporation All rights reserved. Published March 2017 Dell believes the information in this publication is accurate as of its publication date. The information is subject to change without notice. THE INFORMATION IN THIS PUBLICATION IS PROVIDED AS-IS. DELL MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND WITH RESPECT TO THE INFORMATION IN THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. USE, COPYING, AND DISTRIBUTION OF ANY DELL SOFTWARE DESCRIBED IN THIS PUBLICATION REQUIRES AN APPLICABLE SOFTWARE LICENSE. Dell, EMC, and other trademarks are trademarks of Dell Inc. or its subsidiaries. Other trademarks may be the property of their respective owners. Published in the USA. EMC Corporation Hopkinton, Massachusetts In North America EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

3 CONTENTS Tables 5 Preface 7 Chapter 1 Introduction 11 Avamar Data Migration Enabler overview...12 The data migration process Avamar system and client plug-in support with ADMe...12 Capabilities and limitations Capabilities...13 Limitations...13 Architectural Overview Data migration methods Rehydration method...14 Archive Method...17 ADMe Administration Web-UI The ADMe CLI interactive menu...18 Chapter 2 Installing and Getting Started with ADMe 19 Installing the ADMe software Getting started with ADMe Environment files...20 Client group definitions Job policy definitions...21 Configuration workflow Chapter 3 Configuring and Using the ADMe software 23 Starting the ADMe Administration Web-UI...24 Configuring environment files...24 Creating environment files...24 Editing environment files Deleting environment files Viewing the environment files summary report Client group configuration Client group naming conventions...26 Client group limitations Creating client groups...27 Deleting a client group...29 Calculating the aggregate client group size Configuring and using job policies...30 Creating standard job policies...30 Editing standard job policies Creating batch job policies...33 Starting job policies on demand Scheduling job policies...34 Cancelling running job policies EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide 3

4 CONTENTS Viewing job policy summaries...37 Viewing job policy logs Viewing job policy history...37 Deleting job policies Retiring job policies...38 Auditing migrations...38 Recovery Recovery of Microsoft Exchange VSS migrated backups Recovering VMware image backups Recovering Oracle RMAN backups...41 Appendix A Configuring tape applications 43 Supported export applications...44 Auto-scheduling directly from EMC NetWorker Appendix B Structured Staging Paths 47 Structured staging paths...48 Non-incremental staging paths...48 Incremental staging paths...48 Appendix C ADMe Command Line Interactive Menu System 51 ADMe CLI interactive menu system...52 Appendix D Environment File and Client Configuration File Definitions 55 Environment File Contents Client configuration file field definitions Appendix E Command Line Interface Syntax 65 ADMe command line options Appendix F Troubleshooting and Best Practices 79 Troubleshooting and best practices for ADMe software EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

5 TABLES Revision history... 7 Typographical conventions... 8 Environment file attributes and descriptons EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide 5

6 TABLES 6 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

7 PREFACE As part of an effort to improve its product lines, EMC periodically releases revisions of its software and hardware. Therefore, some functions described in this document might not be supported by all versions of the software or hardware currently in use. The product release notes provide the most up-to-date information on product features. Contact your EMC technical support professional if a product does not function properly or does not function as described in this document. Note This document was accurate at publication time. Go to EMC Online Support ( support.emc.com) to ensure that you are using the latest version of this document. Purpose This guide describes how to configure, administer, monitor, and maintain the Avamar system. Audience The information in this guide is primarily intended for system administrators who are responsible for maintaining servers and clients on a network, as well as operators who monitor daily backups and storage devices. Revision history The following table presents the revision history of this document. Table 1 Revision history Revision Date Description 01 December, 2016 DA release of Avamar March, 2017 GA release of Avamar 7.4 Service Pack 1 Related documentation The following EMC publications provide additional information: Compatibility and Interoperability Matrix Release Notes All EMC Avamar client and plug-in user guides EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide 7

8 PREFACE Special notice conventions used in this document EMC uses the following conventions to alert the reader to particular information. NOTICE The Notice convention emphasizes important information about the current topic. Note The Note convention addresses specific information that is related to the current topic. Typographical conventions In this document, EMC uses the typographical conventions that are shown in the following table. Table 2 Typographical conventions Convention Example Description Bold typeface Click More Options. Use for names of interface elements, such as names of windows, dialog boxes, buttons, fields, tab names, key names, and menu paths (what a user specifically selects or clicks). Italic typeface Monospace font Monospace font with italic typeface Monospace font with bold typeface Square brackets EMC Avamar Administration Guide Event Type = INFORMATION Event Severity = OK Event Summary = New group created Type Avamar_server, where Avamar_server is the DNS name or IP address of the Avamar server. Type yes. [--domain=string(/)] --name=string Use for full titles of publications that are referenced in text. Use for: System code System output, such as an error message or script Pathnames, file names, prompts, and syntax Commands and options Use for variables. Use for user input. Square brackets enclose optional values. Vertical bar [--domain=string(/)] --name=string Vertical bar indicates alternate selections - the bar means or. 8 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

9 PREFACE Table 2 Typographical conventions (continued) Convention Example Description Braces {[--domain=string(/)] --name=string} Braces enclose content that the user must specify. Ellipses valid hfs... Ellipses indicate nonessential information that is omitted from the example. Where to get help The Avamar support page provides access to licensing information, product documentation, advisories, and downloads, as well as how-to and troubleshooting information. This information may enable you to resolve a product issue before you contact EMC Customer Support. To access the Avamar support page: 1. Go to 2. Type a product name in the Find a Product box. 3. Select the product from the list that appears. 4. Click the arrow next to the Find a Product box. 5. (Optional) Add the product to the My Products list by clicking Add to my products in the upper right corner of the Support by Product page. Documentation The Avamar product documentation provides a comprehensive set of feature overview, operational task, and technical reference information. Review the following documents to supplement the information in product administration and user guides: Release notes provide an overview of new features and known limitations for a release. Technical notes provide technical details about specific product features, including step-by-step tasks, where necessary. White papers provide an in-depth technical perspective of a product or products as applied to critical business issues or requirements. Knowledgebase The EMC Knowledgebase contains applicable solutions that you can search for either by solution number (for example, esgxxxxxx) or by keyword. To search the EMC Knowledgebase: 1. Click Search at the top of the page. 2. Type either the solution number or keywords in the search box. 3. (Optional) Limit the search to specific products by typing a product name in the Scope by product box and then selecting the product from the list that appears. 4. Select Knowledgebase from the Scope by resource list. 5. (Optional) Specify advanced options by clicking Advanced options and specifying values in the available fields. 6. Click Search. EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide 9

10 PREFACE Online communities Go to EMC Community Network at for peer contacts, conversations, and content on product support and solutions. Interactively engage online with customers, partners, and certified professionals for all EMC products. Live chat To engage EMC Customer Support by using live interactive chat, click Join Live Chat on the Service Center panel of the Avamar support page. Service Requests For in-depth help from EMC Customer Support, submit a service request by clicking Create Service Requests on the Service Center panel of the Avamar support page. Note To open a service request, you must have a valid support agreement. Contact an EMC sales representative for details about obtaining a valid support agreement or with questions about an account. To review an open service request, click the Service Center link on the Service Center panel, and then click View and manage service requests. Enhancing support EMC recommends that you enable ConnectEMC and Home on all Avamar systems: ConnectEMC automatically generates service requests for high priority events. Home sends configuration, capacity, and general system information to EMC Customer Support. Comments and suggestions Comments and suggestions help EMC to continue to improve the accuracy, organization, and overall quality of the user publications. Send comments and suggestions about this document to DPAD.Doc.Feedback@emc.com. Please include the following information: Product name and version Document name, part number, and revision (for example, 01) Page numbers Other details to help address documentation issues 10 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

11 CHAPTER 1 Introduction This chapter includes the following topics: Avamar Data Migration Enabler overview Capabilities and limitations...13 Architectural Overview Data migration methods...14 ADMe Administration Web-UI...18 The ADMe CLI interactive menu Introduction 11

12 Introduction Avamar Data Migration Enabler overview The data migration process Avamar Data Migration Enabler (ADMe) is a software tool for migrating Avamar backup data to a different storage medium, such as tape, cloud storage, or stand alone USB drives. Backup data selected for migration is rehydrated to a staging server, where it can be staged to the various storage types, or export to tape using a tape backup product such as EMC NetWorker. For recovery, migrated files can then be retrieved from storage or restored from tape using the backup application's native recovery tools. With ADMe, the migration process is automated via user-defined migration policies, which are used to select the appropriate Avamar backup for migration. The selected backup is then rehydrated by performing a redirected recovery of the data from Avamar to a structured directory tree on the allocated staging disk area. Optionally, automated export to tape of the rehydrated data can be invoked using a predefined tape policy within the tape backup application. When the data is being written to tape, the tape backup application is responsible for assigning appropriate retention periods and a cataloging mechanism that allow for browsable file-level recoveries. Their recoveries can be written to the ADMe staging server, or to any compatible client activated to the tape backup application. Using the same migration process, ADMe can also be used to export Avamar backups to another Avamar system that uses a Data Domain system for its back-end storage. This allows backups stored in an Avamar GSAN-based storage node to be migrated to a target Avamar/Data Domain system. The resulting migrations are browsable using the target Avamar recovery browser to perform file level recoveries. If the Avamar backups are staged to cloud-enabled storage, the data is then stored in rehydrated format in the cloud and can be recovered using the cloud storage provider's mechanisms for recovery. If the cloud-enabled storage is access through a gateway that supports data deduplication, then the rehydrated data can be stored in deduplicated format in the cloud. Avamar system and client plug-in support with ADMe ADMe supports all Avamar system types including single or multi-node systems, Avamar Virtual Edition, replication sources and targets, and Avamar with Data Domain. The ADMe migration software supports the Avamar filesystem plug-in on all supported filesystems as well as plug-ins for virtual machines in both VMware and HyperV environments, the NDMP plug-in, and all supported database application plug-ins such as the Microsoft Exchange and Oracle plug-ins. An Avamar archival format, PAX, is available, which allows the data to be stored in a native Avamar format which can then be re-imported. There are several limitations on the use of the Avamar archival format; Limitations of the archive method on page 17 addresses these limitations. While various tape backup applications can be used with the ADMe workflow, only EMC Networker provides the full chain-of-custody details associated with an end-toend migration. 12 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

13 Introduction Capabilities and limitations Capabilities Limitations The following capabilities and limitations apply to the ADMe migration tool: Enables exporting Avamar backup data to tape or cloud storage for long-term retention, lowering overall cost of the storage consumed by Avamar itself. Creates transportable media such as tape or standalone USB hard drives for offsite storage. Leverages existing tape applications and associated infrastructure. A variety of compatible servers and operating systems can be used as a staging servers. Supports multiple export use cases at the same time, for example, to both tape and cloud storage. Single point of management for Avamar data migration. No dependency on Avamar when recovering from tape or cloud, except for Oracle and PAX file archives. Tape backups can be cataloged and browsable for file-level recovery. Supports migration of backups to public or private cloud storage. Supports multiple staging servers concurrently, including when more than one thread per staging server is configured. Supports all Avamar system types. Uses a web-based graphical interface, as well as a command-line based menudriven interface. Provides notifications for significant migration activity. Event and job logs capture status of all significant migration activity. Supports incremental and non-incremental staging strategies. Supports most Avamar plug-in types. Has a wide array of backup selection filters to automatically select backups for migration. Provides an audit mechanism to track critical path components and chain-ofcustody. Able to migrate GSAN based backups from one Avamar system to an Avamar server that uses Data Domain as its backend storage, storing the backup data on the Data Domain. Migrations can take a significant amount of time depending on the data size and file counts involved; therefore while your backup software may perform frequent backups (such as on a daily basis), it is not practical to migrate daily backup with ADMe. Capabilities and limitations 13

14 Introduction Architectural Overview When exporting migrated backups to physical tape, a robotic tape library is required as part of the tape backup infrastructure. When migrating backups to the cloud, the upload bandwidth and latency to the cloud provider will have a direct impact on migration throughput. ADMe should not be leveraged to reduce required Avamar capacity. The proprietary Avamar archive approach using a PAX file should not be used when an Avamar metadata node is being used in conjunction with Data Domain backend systems, because importing the PAX file consumes GSAN capacity which cannot be reclaimed for future metadata, even after the imported backup has been deleted. Basic ADMe architecture begins with the data flow from Avamar to the staging server disks, and from there to an optional export application such as a backup-to-tape application. Once client group's backups have been staged, an export to tape can be invoked automatically using a predefined tape policy definition on the backup-to-tape application, which backs up only the staging path destination. If multiple staging servers and staging threads are being used and multiple export methods are available, they can all be used concurrently, which improves the overall aggregate throughput. Data migration methods Rehydration method The Avamar Data Migration Enabler provides two methods for migrating data: Rehydration method Archive Method The rehydration method of data migration performs a directed recovery of Avamar backups to a staging server, where the hydrated data is available for export to the cloud or onto tape. With the rehydration method, a backup of the migrated data to tape will result in all files being in their original rehydrated state and a catalog provided by the tape 14 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

15 Introduction application. Using the tape application's standard recovery mechanisms, file browsing is available and files can be recovered directly from the tape to any available and compatible client using the mechanisms, and subject to the requirements of, the tape application. Similarly, migrated data can be exported to the cloud using cloud-based tools, like you would for normal filesystem data. With the hydration method, there is no dependency on Avamar being present in the environment for recovery of the files from tape or the cloud other than with the Oracle plug-in, making it the most effective approach for long term data retention. Staging server requirements vary based on the rehydration approach used. For complete data rehydration, it is preferable to stage backups of Microsoft Windows data to a staging server on a Windows host, in order to maintain Windows-specific file attributes. You should avoid staging UNIX/LINUX backups to a Windows-based staging server because of case-sensitivity issues for filenames and directories Note Full versus incremental migrations Staging directory structure Non-Windows based backups can be staged to any UNIX or LINUX platform and operating system, although if the file system involved supports an expanded set of file attributes that you want to preserve, you should use that operating system for the staging server. Migrations can either be full or incremental. Full migrations represent the recovery of a complete backup, which, after exporting, will be deleted from the staging server. When using the full migration mechanism, the staging server's disk capacity must at a minimum be large enough to hold the largest individual backup involved. Large file systems and NDMP backups can perform migrations more quickly when using the incremental method, in which staged data is not deleted from the staging server but is retained so that future migrations only needs to stage files that are new or have changed since the previous migrated backup, rather than staging the entire backup. However, this approach requires more staging disk capacity. After the initial full migration, all subsequent migration sessions leverage the time stamps from the previously staged files to determine whether a file needs to be restaged or not. If the backup being staged contains the same or older version of a file that has already been staged, the file will not be migrated, thus increasing the effective staging throughput. When staging is performed incrementally, an associated tape backup application used to export the data can also leverage an incremental tape backup mechanism to export the data. Incrementally staged data can also be accessed directly from the staging disk area at any time. Considering migrations are usually performed on a monthly basis the tape policy can be configured for a full backup once or twice per year in January for example and incremental the rest of the year which would be the equivalent of less than a traditional two week backup cycle. Migrated data on the staging server uses a user-defined path along with a structured directory tree which includes the client name as the name of the folder containing the data for each migrated client backup. If the full migration method is used, the directory tree also includes a folder named after the original Avamar backup date and backup id, using the format yyyy-mm-dd-buid#. Incremental migrations do not include this point in time folder, but a incremental status file showing the dates of the last migration is provided and stored on the staging disk. Rehydration method 15

16 Introduction Migrations of Microsoft backups Staging server requirements vary based on the rehydration approach used. For true rehydration, it is preferable to stage backups of Microsoft Windows data to a Windows-based staging server, in order to maintain Windows-specific file attributes. You should avoid staging UNIX/LINUX backups to a Windows-based staging server because of case-sensitivity issues for filenames and directories Note If performing a full migration to a Microsoft Windows-based operating system, the quick format command is used to delete data from the staging drive. Therefore, it is a prerequisite for the staging drive letter be dedicated to ADMe use. Software requirements for Windows-based staging servers At a minimum, the staging server must have the Avamar filesystem plug-in installed. The following plug-ins have additional requirements: Sharepoint VSS The staging server must also have the Sharepoint VSS plug-in installed. HyperV VSS The staging server must also have both the HyperV VSS plug-in and Microsoft HyperV software installed. Exchange VSS The staging server must also have both the Exchange VSS plug-in and Microsoft Exchange software with the with the Mailbox role enabled. Note For Exchange backups, the logs are applied during the staging process, resulting in only the EDB files being archived. Migration of UNIX and Linux-based backup data Using cloud storage for long term retention Non-Windows based backups can be staged to any UNIX or LINUX platform and operating system, although if the file system involved supports an expanded set of file attributes that you want to preserve you should use that operating system for the staging server. The staging server must have the Avamar filesystem plug-in installed on it. No additional plug-ins are required on UNIX and Linux-based staging servers, regardless of the type of plug-in used to create the backups that are being staged. Object based storage such as EMC ECS, EMC ATMOS, or other cloud providers can be used as a staging target, resulting in a simple approach to migrating large amounts of backup data to Cloud storage without impacting or disrupting your local Avamar backup environment and possibly eliminating the need for a tape application. Cloud staging targets can be enabled via cloud gateway technologies such as EMC ECS Gateway, EMC Cloud Array, and similar tools. Once data is uploaded to the cloud, the gateway tools or other methodologies can be used to access and browse the exported data if it is in rehydrated format. Typically when using a cloud gateway, no local staging disk is required. However, some gateways rely on a local disk cache to act as a temporary buffer to help speed up the overall process. 16 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

17 Introduction When using the incremental migration method with EMC ECS Gateway, EMC Cloud Array, and other gateways, the process results in a single instance of a file being maintained, rather than requiring that multiple instances of files be stored as is the case with the full migration method. This concept enhances throughput and reduces the cloud storage consumed. Archive Method The archive method for data migration creates an archive file that contains rehydrated backup files which have been streamed into an Avamar compatible PAX file. Because this process is common to all Avamar plug-ins, a PAX file should be able to support any Avamar plug-in type, and contains both data files and their related Avamar metadata objects. Recovery from a PAX file requires it first be imported back into an Avamar GSAN in order to make it usable for recovery. As a result, there may be potential versioning compatibility issues over time, especially when dealing with application plug-ins. Therefore the archive method is not a good choice for long-term retention of migrated data. An archive file normally is slightly larger in size than its corresponding fully rehydrated migration, but can optionally be compressed, resulting in a file size reduction usually of about 50%. Compressing an archive file does increase the staging time as it is CPU intensive, but this time can be regained if exporting to tape, because of the reduced file size and lack of indexing overhead since only one file is involved. However, this also implies there is no indexing for individual files, nor any view of individual files from the tape application's recovery browser, therefore there is no ability to select individual files for recovery prior to the data being imported to an Avamar system. To alleviate concerns about having no catalog or logical view of files when writing to an archive file, you can optionally configure ADMe to create a metadata file containing pertinent details of the archive content, including file names, paths, and time stamps. This associated metadata file can be retrieved directly from tape or cloud prior to recovery, and its contents manually reviewed to confirm whether the desired file names with appropriate timestamps are present within the actual archive prior to retrieving the archive itself. Limitations of the archive method The following limitations of the archive method should be considered when choosing this as your migration method. Archive PAX files can only be created on a Linux staging server. There is no logical catalog in the tape application indexing the PAX file content. There is a 100% dependency on the need for an Avamar GSAN being available in order to recover files from the PAX file. As a result, this method is not a good choice for long term retention. An import of the entire PAX file is always required to recover any of its logical file contents. Potential operating system limitations associated with a large PAX file size; similarly there are practical implications with importing very large files. Archive Method 17

18 Introduction ADMe Administration Web-UI The ADMe web-based administration user interface is a graphical management console used to configure and monitor migration policies and administer an ADMe implementation. The ADMe Job-Activity panel will appear when you log in to the administration interface, providing at-a-glance information about the current or most recent run status of each job policy. When used for exporting to tape with EMC NetWorker, export details will be included for each job entry. Additional tabs are available for viewing and configuring environment files, clients, and migration job policies, and for accessing the audit mechanism used to track chain of custody of migration activities. The ADMe CLI interactive menu ADMe also includes a command line based interactive menuing system for configuration and administration of your ADMe implementation. The ADMe CLI interactive menu system is an interactive user interface that can be used to initiate and determine job policy status, view and modify configuration files, view event and job logs, and view selected backup metrics. 18 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

19 CHAPTER 2 Installing and Getting Started with ADMe The chapter includes the following topics: Installing the ADMe software Getting started with ADMe Installing and Getting Started with ADMe 19

20 Installing and Getting Started with ADMe Installing the ADMe software ADMe is distributed as a tar file and installed onto an Avamar utility node, single-node system, or an Avamar Virtual Edition. Before you begin ADMe requires Korn shell for functionality and mailx for mail notifications. Ensure that these are installed on the utility node prior to performing this task. Additionally, the Avamar system must not be in read-only mode during the installation. Procedure 1. Downloaded the latest ADMe installation files from products/779_avamar-data-migration-enabler. The files are named adme-tar-version.tar and adme-version.war, where version is the version of the ADMe software being installed. 2. Log in as root to the Avamar utility node or single-node system. 3. Create a directory on the utility node or single-node system where the installation files will be copied: mkdir /atoadmin 4. Transfer adme-tar-version.tar and adme-version.war to the / atoadmin directory. 5. Rename adme-version.war to adme.war: mv adme-version.war adme.war 6. Unpack the adme-version.tar file: tar -C/ -xf adme-version.tar 7. Execute the installation: adme -upgrade 8. Verify that the administration web service is active: adme -gui status After you finish Getting started with ADMe If you will be using EMC NetWorker to export migrated backups to tape, you must also install the NetWorker 64-bit Linux client agent onto the utility node or stand-alone server. NetWorker documentation contains installation information. The following section describes information you will need to configure your ADMe environment and the process involved in configuration. Environment files Environment files are used to define your staging server infrastructure, which includes the staging server name. They can also optionally include information about the tape application involved in exporting the migrated data. These details are stored in environment files as a series of variables including the staging server names, the staging server's operating system type, Avamar and tape 20 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

21 Installing and Getting Started with ADMe Client group definitions Job policy definitions Configuration workflow application binary paths, mail notification address list and several other operational parameters. There must be one environment file defined per unique staging server or staging server thread name being used, therefore a staging server name cannot be used in more than one environment file. Staging servers must be activated to Avamar, ideally to the /ADME domain, before they can be added to an environment file. Environment files are named using a numbering scheme; the first environment file you create is referenced using Environment file-1, the second is Environment file-2, etc. After creating environment files, you need to establish client groups, which define and segment the migration workload into manageable sizes. The number of clients that a client group can contain is determined only by the practical limitations associated with their overall aggregate data size. Even if only one client is involved, it must be placed in a client group. The same client cannot be added more than one time to the same client group, but a client can be added to any number of different client groups. Clients within a group should be based on a given plug-in type (for example, a group containing Exchange clients should contain only Exchange clients). The final step is to define job policies, which are used to control all aspects of the endto-end migration process. Job policies contain key references such as the name of the client group on which the policy will act, the staging server to be used (as referenced by the specified environment file), and the backup selection and migration criteria that ADMe will use to automatically determine which backups should be migrated. The following diagram depicts the configuration sequence for ADMe and how the components interact within a job policy definition to accomplish a migration. Client group definitions 21

22 Installing and Getting Started with ADMe 22 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

23 CHAPTER 3 Configuring and Using the ADMe software This chapter includes the following topics: Starting the ADMe Administration Web-UI Configuring environment files Client group configuration...26 Configuring and using job policies Auditing migrations Recovery...39 Configuring and Using the ADMe software 23

24 Configuring and Using the ADMe software Starting the ADMe Administration Web-UI Use this procedure to start the ADMe Administration Web-UI. Procedure 1. Open a web browser and enter the following URL: The Avamar ADME UI login page appears. 2. Type the username of the Avamar administrator user account in the Username field. 3. Type the password of the Avamar administrator user account in the Password field. 4. Click Login. Results The ADMe Job-Activity panel will appear when you log in to the administration interface, providing at-a-glance information about the current or most recent run status of each job policy. Configuring environment files Creating environment files The first step in the process of configuring ADMe migrations is to create environment files. Environment files are used to define your staging server infrastructure, which includes the staging server name. They can also optionally include information about the tape application involved in exporting the migrated data. Use this procedure to create environment files. Before you begin Before creating environment files in ADMe, staging server must be activated to the Avamar server, ideally to the /ADME domain. Additionally, if a tape application will be used to export the migrated data, the staging server must also be a valid client of your tape application. Follow regular product documentation for installing and activating clients using their appropriate file system agents. Procedure 1. From the ADMe Administration Web-UI, click Environment. The Environment tab is displayed. 2. From the toolbar on the upper right, click the + tool. The Add Environment window opens. 3. Complete the following information: a. For Export Server Hostname, enter the DNS name of your tape application master server, or in the case of Avamar GSAN to Data Domain migration, the Avamar target host name. 24 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

25 Configuring and Using the ADMe software b. For Export Method, select the appropriate method: For tape-based exports, select the appropriate tape application. For cloud-based export, select Cloud-GW. For migration of Avamar GSAN to Data Domain migration, select avamar. c. For Staging Server, click Select Staging Server and select the appropriate staging server. Staging Server OS Type is detected automatically based on the Staging Server selection. d. For Mail To Address, enter a list of destination addresses, separated by a + character, to which all notifications from this environment will be sent. e. For Error Mail To Address, enter a list of destination addresses, separated by a + character, to which only error notifications from this environment will be sent. Note After configuration of the first environment file, the information you enter into several of these fields propagate to subsequent environment files by default, but these default selections can be edited when subsequent environment files are created or modified. The fields that propagate to subsequent environment files are: Export Server Hostname Export Method Mail To Address Error Mail To Address 4. Click Add to add the environment. Results Editing environment files Environment files are named using a numbering scheme; the first environment file you create is referenced as Environment file-1, the second is Environment file-2, etc. Use this procedure to edit existing environment files. Procedure 1. From the ADMe Administration Web-UI, click Environment. The Environment tab is displayed. 2. Edit information for an environment file directly in the table view by clicking the cell that contains the information to be edited. The following fields are editable: Export Server Hostname Export Method Mail Address Editing environment files 25

26 Configuring and Using the ADMe software Error-Mail Address Deleting environment files Use this procedure to delete existing environment files. Procedure 1. From the ADMe Administration Web-UI, click Environment. The Environment tab is displayed. 2. Select the environment file to be deleted. 3. Click the - tool to delete the environment file. Viewing the environment files summary report The environment files summary report provides a view of all configured environment files and a list of jobs that have been assigned to each environment. Procedure 1. From the ADMe Administration Web-UI, click Environment. The Environment tab is displayed. 2. Click the View Summary tool. Client group configuration Client group naming conventions The Environment Summary window is displayed. 3. Click Close to close the Environment Summary window Client groups segment the migration workflow into managable sizes and can be used to group clients of the same plug-in type. Using a meaningful naming convention for client groups can help organize the migration workflow. Use generic but descriptive names, followed by a numeric suffix; the numeric suffix allows for multiple groups of the same type using the same generic name. This scheme allows for clients to be added or removed from a group without resulting in the group name becoming irrelevant. Examples of group names: winfs1, winfs2, winfs3 Descriptive names for clients that have Windows filesystem backups. lnxnfs1, lnxfs2, lnxfs3 Descriptive names for clients that have Linux filesystem backups. exchvss1, exchvss2, exchvss3 Descriptive names for clients that have Microsoft Exchange backups. sql1, sql2, sql3 Descriptive names for clients that have Microsoft SQL backups. ora1, ora2, ora3 26 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

27 Configuring and Using the ADMe software Client group limitations Creating client groups Descriptive names for clients that have Oracle backups. vmimg1, vmimg2, vmimg3 Descriptive names for clients that have VMware image backups. The following are limitations and considerations when creating client groups: There are no hard limits to the number of client groups permitted. There is no hard limit to the number of clients that can be added to a group, other than the practical impact of the aggregate workload size involved. A client can be added to any number of groups but can be added only once to a single group. The staging path location can vary between clients in a group, but generally is not configured this way. A number of the selection criteria filters, such as the Label Name Filter and the Retention Tag Value, can vary between clients in a group. The Staging Server OS setting must be the same across all clients in a group. The Monthly Tape Policy, Yearly Tape Policy, and Server Tape Policy, if used, must be the same across all clients in a group. Procedure 1. From the ADMe Administration Web-UI, click Client Config. The Client Config tab is displayed. 2. From the toolbar at the top of the Client Groups panel on the left side of the screen, click the + tool. The Create Client Group window opens. 3. Complete the following: a. For Group Name, enter a group name, limited to eight characters. Avoid using special characters such as \$&#*(){}. b. The Label Name Filter is a filtering mechanism used by the ADMe software to determine which client backup should be migrated when one more than one backup of the same type exists on a given day. Enter any unique string value present in the name field which can be viewed from the Avamar recovery interface. c. The Retention Tag Value is a filtering mechanism used by the ADMe software to limit the backups selected for migration based on a specific retention tag value. Allowed values are none, daily, weekly, monthly, yearly, and n/a if not used. This value can be set on a per-client basis for clients in the same group. d. For Exclude Files, specify one or more folders or files that will be excluded from the staging process. Multiple names can be specified, separated by a pipe character. e. For Include Files, specify one or more folders or files that will be included in the staging process. All other folders/files will be excluded from the staging process. Multiple names can be specified, separated by a pipe character. Client group limitations 27

28 Configuring and Using the ADMe software Adding clients to a client group f. For Staging Path, specify the path on the staging server filesystem where data will be written during migration. Each client within the group can have a different staging path. Never stage data to the root directory of any mount point, always specify at least one folder. g. For Staging Server OS, select the appropriate operating system. h. For Destination Path Suffix, enter a staging path suffix which will be inserted between the final staging path and the actual data files migrated. i. For Monthly Tape Policy, enter a tape policy name to be called automatically. For EMC NetWorker, this represents the NetWorker group name. j. For Yearly Tape Policy, select a tape policy name that will be used instead of the Monthly Tape Policy for backups performed based on the setting of the End Of Year variable in the environment file. For example, if the End Of Year variable is set to 01,06, then this field's value will be used in the months of January and June, while the value for Monthly Tape Policy will be used in the other months. This facilitates having an automated end-ofyear backup where the retention value on tape may differ from the other months of the year. k. For Server Tape Policy, enter a tape policy name to be initiated from the tape master server rather than the more common method of using a client initiated tape backup from a staging server client. 4. Click Add Group to create the group. Use this procedure to add clients to a client group. Clients should not be added to the same group more than once but can be present in multiple groups. Before you begin Note When clients are added to a group, they will automatically inherit the contents from the previous client entry in the group. As a result, you should make sure when adding clients that these settings are appropriate for the newly added client. Procedure 1. From the ADMe Administration Web-UI, click Client Config. The Client Config tab is displayed. 2. Highlight an existing client group name in the left pane. The existing client entries for the selected client group will be displayed in the right pane. 3. From the toolbar on the upper right, click the + tool. The Add Clients window opens. 4. In the Add Clients window, navigate the client tree and select each client to be added to the group. 5. Click Select when complete to add the selected clients to the group. Results 28 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

29 Configuring and Using the ADMe software Deleting clients from a client group Use this procedure to delete clients from a client group Before you begin Procedure 1. From the ADMe Administration Web-UI, click Client Config. The Client Config tab is displayed. 2. Highlight an existing client group name in the left pane. The existing client entries for the selected client group will be displayed in the right pane. 3. Select an existing client entry from the table in the right pane. Multiple clients can be selected using the Ctrl Key 4. From the toolbar on the upper right, click the - tool. 5. Click Yes to confirm the deletion. Results Deleting a client group Use this procedure to delete a client group Procedure 1. From the ADMe Administration Web-UI, click Client Config. The Client Config tab is displayed. 2. Select a client group from the list in the left pane. 3. From the toolbar at the top of the Client Groups panel on the left side of the screen, click the - tool. 4. Click Yes to confirm the deletion. Calculating the aggregate client group size Calculate the aggregate size of a client group using this procedure. Procedure 1. From the ADMe Administration Web-UI, click Client Config. The Client Config tab is displayed. 2. Select a client group from the list in the left pane. 3. From the toolbar on the upper right, click the Check Group size tool. The Group Size Check window is displayed. 4. Click Close to close the Group Size Check window. Deleting a client group 29

30 Configuring and Using the ADMe software Configuring and using job policies Job policies coordinate the end-to-end migration process, controlling each phase of workflow. They facilitate repeated migrations on a scheduled basis or can be initiated on demand when required. Two policy types exist: Creating standard job policies Standard policies perform the migration workflow. Batch policies are used to initiate a series of standard policies in sequence. Standard job policies perform the migration process, while batch policies are used to simplify scheduling by acting on a series of standard policies at a time. Use this procedure to create a standard job policy. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. From the toolbar on the upper right, click the + tool. The Backup Selection Criteria window displays. 4. Complete the following information, if required: a. For Job Name, enter a job name, limited to eight characters. Avoid using special characters such as \$&#*(){}. b. For Environment-#, select the number of the environment file that defines the staging server that this policy will use. c. For Group, select the client group on which this policy will act. d. For Plug-in Type, select the appropriate plug-in. This information is visible in the Avamar recovery interface for verification. 30 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

31 Configuring and Using the ADMe software Note When using a database application plug-in, you should only migrate full backups. Use the Label/Name Field String, below, to refine your filter based on a unique string value associated with the backup to make sure that you are only filtering for full backups. e. For First or Last, select whether to use the first or last backup in the data range. f. For Retention Tag, select a retention tag to limit the selection of backups for migration based on their retention tag. g. For Days of the Week, select whether to limit the selection of backups for migration to only backups that take place on a particular day of the week. h. The Week Num field works with the Days of the Week field to limit the selection of backups for migration to only the backup from a particular week during the month. i. For Include Backup, select whether to include on-demand backups, as well as scheduled backups, when determining which backups should be migrated. COD are on-demand backups initiated from the client. MOD are on-demand backups initiated from the Avamar Administrator. NAH are naked Ad Hoc backups. j. For Label/Name Field String, select whether to limit the selection of backup to backups whose label field includes the string listed here. The label field is visible in the Avamar recovery interface. Normally the label field contains the Avamar group and schedule name strings, but can also be a user-defined label value. 5. Click Next. The Data MIgration Criteria window is displays. 6. Complete the following information: a. For Data Migration Action, select whether this job should perform only the migration staging, only the tape backup phase or both. The default behavior is to perform the migration followed by the initiation of a tape backup of the staged data. There are circumstances where it may be desirable to have a job perform only the migration staging or only the tape backup phase, but not both. Creating standard job policies 31

32 Configuring and Using the ADMe software b. Do not change the selection for Recovery Method. 32 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide c. For Staging Method, determine whether to stage the backup data nonincrementally or incrementally. The default is to use full (non-incremental). d. For As an Archive, determine whether the job policy should store the backup data as a tar file or an Avamar archive (PAX). e. For Purge Migrated Backups-ID(s), determine whether to automatically purge migrated backup IDs from Avamar. Lower case -purge verifies an export/tape backup has completed successfully prior to purging the backup ID from Avamar. Upper case -PURGE will purge the backup ID once it has been successfully staged only. This option should be used with caution and only when historical data is being migrated. f. The Cloud Expiry Date provides a visual and controlled expiry date used when migrating data to cloud storage. The visual aspect enables you to view and manually expire folders according to its planned retention period with or without the use of ADMe. It works based on the use of a key word EXPIRE being present anywhere within the user defined staging path that was used when creating a client group. Then, based on the following Cloud Expire options, the keyword is converted automatically into a YYYY-MM-DD format corresponding to the time period you chose. The keyword of EXPIRE can be preceded with a string value such as Expire-Date- or Purge-Date. Expire Fixed Date specifies a planned expiry data using a calendar provided. By Time Period allows ADMe to calculate the planned expiry date based on the number of days, weeks, months, or years chosen. Select Round to round of the time period to the 1 st of the last+1 calculated month. Could Data Purge performs an automated deletion process against eligible expiry folders described above. When using this option, you must assign a client group which should be one normally staged to the staging server defined to the environment number specified and the selection flags should be left empty. There may be conditions where Windows permissions prevent the deletion process from being 100% successful and manual intervention may be required to complete the deletion process. If this condition exists it will be reported in the job log. g. For Staging Path Suffix, determine whether to insert a folder name immediately prior to where the backup data is staged, and enter the path h. For Post Migration User Script, enter a user-defined script to be executed immediately after a tape backup has completed successfully. The script must reside on the staging server in the Avamar agent install folder under /etc/ scripts. For Windows, the script file name must conform to 8.3 filename format. A common use case for this is to ensure staged data can be is removed reliably from a Windows staging server. i. For Folder Include or Exclude, determine whether to include or exclude top level folders from a backup to limit the migration workload, and enter the folder name to be included or excluded. Selecting -data will include the folder, and selecting -xdata will exclude it. The Web-UI supports a single call but multiple calls may be made by manually entering them into the Migration Command area. When using this option, it will apply to the entire client group. If your requirement is for only a specific client within a group, refer to Client Config tab, where embedded filters can be applied at the client level.

33 Configuring and Using the ADMe software Editing standard job policies j. For Export Policy Name Override, the default monthly tape policy name is defined within a client group, however, there may be situations where a migration of a given group requires the use of more than one tape policy. As an alternative to creating another client group, you can create another job policy and pass a different tape policy to it, overriding the one defined within the client group itself. k. For Include Meta Data Listing, determine whether to create a catalog-like listing of the contents of an export which can be reviewed prior to retrieving the actual export from tape or cloud storage. This option is intended to be used when staging to an archive file such as a tar or PAX file. When using this option for other than an archive, you must override the Recovery Method option to use -avtar. l. For Custom Avtar Flags, pass any additional specific avtar flags used to influence a backup recovery, rather than placing it into an avtar.cmd file on the staging server. For example, by default a Windows file system restore does not include files with the system attribute set therefore, if you require these files you can pass the required avatar flag here restoresystem. Do not include any preceding hyphens. m. The Migration Command field echoes the various options chosen where their syntax can be viewed, and can also be used to include options not available within the Web-UI. 7. Click Create Job. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. Edit information for a standard job policy directly in the table view by clicking the cell that contains the information to be edited. The following fields are editable: Backup Selection Criteria Backup Migration Criteria Client Group name Thread Number Note To edit other fields, use the CLI interactive menu system. Creating batch job policies Use this procedure to create batch job policies A batch job policy behaves like a script that calls a series of standard job policies sequentially until all have been processed. They simplify scheduling, as only the batch Editing standard job policies 33

34 Configuring and Using the ADMe software policy needs to be scheduled, which eliminates conflicts resulting from environment file locks that will prevent a policy from starting if another policy is already using the environment. Since the assigned polices are processed sequentially, a batch policy should only include standard policies that use the same environment file. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. From the toolbar on the upper right, click the Add Batch Job tool. The Create Batch Job window displays. 4. Complete the following information: a. For Job Name, enter a name for the batch job. b. For Fail Action, determine if the batch job should stop executing, or continue processing additional scheduled jobs after one scheduled job fails. c. For Standard Job Names, select the standard jobs policies that should be included in this batch policy job. To multi-select standard job policies, press and hold the Ctrl key and highlight each job policy that should be included. 5. Click Create Batch Job. Starting job policies on demand Scheduling job policies Use this procedure to start job policies on demand. Either standard or batch policies can be started using this procedure. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. Note This action can also be performed from the Job Policy tab. 2. Highlight the row of the job policy you want to start. 3. Right-click and select Start Job. Use this procedure to schedule job policies. Either standard or batch policies can be started using this procedure. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 34 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

35 Configuring and Using the ADMe software 2. Click Job-Policies. The Job Policies tab is displayed. 3. Highlight the row of the job policy you want to schedule. 4. Right-click and select Schedule-Enable.. The Cron Parameters window appears. 5. Complete the following information: a. For Time, enter or select the time of the day that the scheduled job will start using an HH:MM format. b. For Date, enter the day of the month that the job will start. Allowed values are 1 to 31. An asterisk means that the job will run on any day of the month as determined by the Month and Day settings, below. Multiple dates can be included, using one of the following formats: d1+d2...+dn; for example a value of 1+15 will run the schedule on the first and fifteenth of each month. d1-d2; for example 1-7 will run the schedule for each of the first seven days of the month. A combination of both mechanisms; for example will run the schedule for each of the first seven days of the month and again on the fifteenth. c. For Month, enter the month of the year that the job will start. Allowed values are 1 through 12. An asterisk means that the job will run each month, assuming the Date value can be fulfilled that month. Multiple months can be included, using one of the following formats: m+m2...+mn; for example a value of will run the schedule on the months of January, April, August, and December only. m1-m2; for example 2-12 will run the schedule every month except January. A combination of both mechanisms; for example will run the schedule for January, February, and March, and again on October, November, and December. d. For Day, enter the day of the week that the job will start. Allowed values are 0 through 6, with 0 corresponding to Sunday. An asterisk means that the job will run on any day of the week as determined by the Date and Month settings, above. Multiple days can be included, using one of the following formats: d1+d2...+dn; for example a value of will run the schedule on Firday, Saturday, and Sunday. m1-m2; for example 1-3 will run the schedule on Monday through Wednesday. A combination of both mechanisms; for example will run the schedule on Monday through Wednesday, and again on Saturday. Scheduling job policies 35

36 Configuring and Using the ADMe software 6. Click Save. Disabling schedules Removing schedules Use this procedure to disable the schedule for a job policy. Either standard or batch policies can be disabled using this procedure. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. Right-click and select Schedule-Disable. 4. Click Yes to confirm that the schedule should be disabled. The schedule can be reeanbled by selecting Schedule-Enable. Use this procedure to remove the schedule for a job policy. After removing a job from the scheduler and you need to reschedule it again in the future, you will need to re-add it to the scheduler. Either standard or batch policies can be disabled using this procedure. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. Right-click and select Schedule-Remove.. The scheduled is immediately deleted. Modifying schedules Use this procedure to modify the schedule for job a policy. Either standard or batch policies can be modified using this procedure. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. Right-click and select Schedule-Modify.. The Cron Parameters window displays. 4. Modify settings in the Cron Parameters window and click Save. 36 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

37 Configuring and Using the ADMe software Cancelling running job policies Use this procedure to cancel running job policies. Procedure Viewing job policy summaries Viewing job policy logs 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Highlight the row of the running job name you want to cancel. 3. Right-click and select Cancel Job. The Cancel Confirmation window appears. This windows explains the implications of cancelling the job. 4. In the Cancel Confirmation window, enter a reason for the cancellation and click Yes. In may take several minutes for the job to cancel. Use this procedure to view the summary of a job policy. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Highlight the row of the job name for which you want to view the summary. 3. Right-click and select View Summary. The Log window appears with the job summary displayed. Use this procedure to view the log for a job policy. Procedure Viewing job policy history 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Highlight the row of the job name for which you want to view the log. 3. Right-click and select View Log. The Log window appears with the job log displayed. Use this procedure to view the history of a job policy. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. Cancelling running job policies 37

38 Configuring and Using the ADMe software Deleting job policies Retiring job policies 2. Highlight the row of the job name for which you want to view the history. 3. Right-click and select View History. The Job History window appears. The history of all past occurrences of the running of the job policy is displayed. To delete a job policy, follow this procedure. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. Highlight the row of the job policy name you want to delete. 4. Click the - tool to delete the job policy. Results Deleting a job deletes the policy definition, its associated runtime log, its historical logs, and removes the job from the scheduler, if configured there. To retire a job policy, follow this procedure. Procedure 1. From the ADMe Administration Web-UI, click Job-Policy Mgr. The Job Policy Manager window is displayed with the Job Activity tab displayed. 2. Click Job-Policies. The Job Policies tab is displayed. 3. Highlight the row of the job policy name you want to retire. 4. Right-click and select Retire Job. Results Auditing migrations Retiring a job retires the policy definition, its associated runtime log, its historical logs, and removes the job from the scheduler, if configured there. A retired job can be reinstated by manually moving these files from the retire folder back to the appropriate folder. The auditing mechanism tracks each migration session performed by ADMe. Procedure 1. From the ADMe Administration Web-UI, click Audit. The Audit tab is displayed and audit records are displayed. 2. For information about each column in the Audit tab, click the Audit Help tool. 38 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

39 Configuring and Using the ADMe software Recovery All recoveries from a migrated backup, whether from tape or cloud storage and regardless of the plug-in type involved, are performed at the file/folder level using the tape applications recovery tools, or copying and pasting from a cloud target. With a tape application, file level recovery can be directed to any target client compatible with the staging server used to perform the migration backup. Database application plug-in type migrations will require additional processing after their files have been recovered from tape or cloud storage to make the data usable to their corresponding database application. The following sections outline the requirements and processes involved. Filesystem data can be recovered at the file level directly to their desired location, or moved to the desired location using operating system tools. NDMP data can be recovered at the file level directly to their desired location, or moved to the desired location using operating system tools. Lotus Notes data can be recovered at the file level directly to their desired location, or moved to the desired location using operating system tools. SQL data is recovered as flat file named f-0 for each database involved in the migration. This file is formatted the same as a regular SQL dump file. Therefore, in order to import it as a working SQL database, follow the same procedure used to import any SQL dump file. Hyper-V VSS backups are recovered as a collection of virtual system files containing the Hyper-V server. These can be copied into their desired location. SharePoint VSS backups are recovered as a collection of flat files including the SQL databases. These recovered files can be copied into their desired location but are not suitable to perform a disaster recovery of a SharePoint farm. Recovery of Microsoft Exchange VSS migrated backups Microsoft Exchange VSS backups are recovered as flat files, resulting in only the EDB files being captured to tape. During the initial staging phase, all associated transaction logs are applied to the EDB files, bringing them up to date with the point-in-time of the original Avamar backup. If the process of applying the transaction logs fails, then the migration itself will not succeed. To recover an entire database, place the EDB file in the folder of the same production database and mount it (dismount the existing database and move the existing files to a safe location first, if needed). Note that with this method, you will lose any data created in the production database after the point of the backup. To recover individual mailboxes/items, you need to establish an RDB pointing it to the temporary location of the recovered EDB file and mount it then use the Microsoft Restor box powershell cmdlet to restore items. Refer to the following links for more detail: Recovery 39

40 Configuring and Using the ADMe software Recovering VMware image backups You can also use EMC ItemPoint of Exchange to mount the EDB file and perform granular recovery of individual mailboxes/items, as described in the for Exchange VSS User Guide. WMware image backups are recovered as flat files. To import them into vcenter you must use the following procedure on a per-image basis. Procedure 1. Recover the appropriate migrated backup files from tape or cloud to a temporary folder by doing one of the following: Mount a datastore to the staging server via CIFS/NFS and direct the tape recovery to a datastore filesystem. (This method is preferred, because it will eliminate subsequent transfer times.) Enable SSH on the ESX server and use sftp to upload the folder where the files were recovered. In the vsphere client, select the datastore on you which want to place the VM and open the Datastore Browser to upload the folder where your renamed files reside. 2. Rename files vm.vmx, virtdisk-flat.vmdk, virtdiskdescriptor.vmdk to match the name to be used by the recovered VMware client. For example, rename the files to restoredvm.vmx, restoredvmflat.vmdk, restoredvm.vmdk, where restoredvm is the named used by the recovered VMware client. 3. Move all the renamed files to a single directory. 4. Review the restoredvm.vmx file in a text editor. Ensure that the hard disk path locations match the renamed files and edit if necessary. For example: Before editing the file reads: scsi0:0.filename = "vmname.vmdk" After editing: scsi0:0.filename = "restoredvm.vmdk" where restoredvm is the named used by the recovered VMware client. 5. Review the restoredvm.vmdk file in a text editor. Ensure that the hard disk path locations match the renamed files and edit if necessary. For example: Before editing the file reads: # Extent description RW VMFS "vmname-flat.vmdk" 40 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

41 Configuring and Using the ADMe software After editing: # Extent description RW VMFS "restoredvm-flat.vmdk" Recovering Oracle RMAN backups where restoredvm is the named used by the recovered VMware client. 6. Register the VM in the Datastore Browser by right-clicking on the recovered.vmx file and selecting Add to inventory. The migration of scheduled RMAN backups result in the backup pieces including the control file being staged as flat files and written to tape or cloud. When recovering these from tape, place the related backup pieces in a common folder. RMAN backups depend on the Oracle SBT library, which backup tools such as Avamar must use. The recovery of an RMAN backup also depends on the SBT library, therefore the flat files as migrated cannot be used by Oracle directly. In order to leverage a migrated RMAN backup, the backup pieces must be re-ingested back into Avamar using a custom flag file to inform Avamar this is an Oracle backup. RMAN recovery procedures can now be used for recovery. Procedure 1. Ensure that a compatible version of the Oracle software is available to establish a test database named DBTS1. 2. Create an appropriate sub-folder to receive the recovered files, for example D: \ADME_restore\11g\DBTS1 3. Recover the migrated backup files to the sub-folder created above (for example, D:\ADME_restore\11g\DBTS1). 4. Create a custom flag file named myflags.txt in D:\ADME_restore. The contents of the myflags.txt file should be: --freezemethod=none --pidname=oracle --pidnum=pidnum --logfile=c:\progra~1\avs\var\admeoracle.log --vardir=c:\progra~1\avs\var --id=username --ap=password --path=/dev/db server.domain --server=avamar server --expires=5 --ddr --ddr-index=1 where: pidnum is 3002 for the Avamar plug-in for Oracle on Microsoft Windows, or 1002 for the Avamar plug-in for Oracle on Linux. username is the username used to log into the Management Console. password is the password used to log into the Management Console. DB server.domain is the Avamar account path to the Oracle client. Recovering Oracle RMAN backups 41

42 Configuring and Using the ADMe software avamar server is the Avamar HFS name. Note --ddr and --ddr-index=1 are required when Data Domain is the intended backup destination. 5. Re-ingest the RMAN backup pieces back into Avamar as a single backup using the avtar command, for example: cd D:\ADME_restore avtar -c --label=adme_test --flagfile=myflags.txt "11g/DBTS1" Results This creates an Avamar backup with the name adme_test. The database is now recoverable. The for Oracle User Guide has information about recovering Oracle databases. 42 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

43 APPENDIX A Configuring tape applications This appendix includes the following topics: Supported export applications Auto-scheduling directly from EMC NetWorker...45 Configuring tape applications 43

44 Configuring tape applications Supported export applications ADMe treats all export applications in a similar fashion, relying on a predefined backup application policy first being established in the backup application and the appropriate syntax being passed to the backup application from the client configuration file field Fld-12 or Fld13. In all cases, an appropriate backup application command line syntax is called by ADMe, using the Fld-12 or Fld-13 contents as arguments to initiate the tape export. Export application EMC Networker Cloud Gateway Arcserv Avamar DD-Migrate Backup Exec Backup Exec 2012 CommVault Dell NetVault HP Data Protector NetBackup TSM CLI command save and savefs N/A ca_backup avtar bemcmd (available on Windows only) bemcmd (available on Windows only) qlogin and qoperation backup (CV requires authentication) nvtrigger and nvjobstart omnib bpbackup dsmc archive When a tape failure occurs, ADMe reports its return code value and any messages returned by the tape application command being called. These are important first steps in determining the cause of a tape failure. All tape applications return a RC=0 for a successful operation with the exception of Backup Exec, which returns RC=1 for success. Additional notes A job log always shows the actual tape command syntax issued from the staging server involved under the Export-Details section. A copy/paste of this command on the staging server can be used to run it manually outside of ADMe for diagnostic purposes. The complete tape script can be viewed on the utility node in the appropriate temp folder based on the environment number being used /atoadmin/tmp/atocfg, atoadmin/tmp/atocfg2, etc. The tape script file name is ADMEtapeout.tab or ADMEtapeout.sh and will be hidden if the job has run to completion. This file will remain until another tape out session is started from this environment file number. A file named autotapeout.stat captures the tape command script output and is available on the staging server under the etc/scripts folder or can be found on the Avamar utility node as a hidden file in the appropriate environment number temp folder as mentioned above. 44 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide

45 Configuring tape applications By default the tape script will be deleted from the staging server upon its completion. To retain a copy of the script on the staging server add four -d debug flags to the Migration_Cmd line in a job profile but only do this when necessary as the multiple debug flags result in a lot of other details being logged. If ADMe is unable to access the contents of the auottapeout.stat file, it cannot determine with certainty the tape backup's completion status. There could be any number of reasons for this situation to exist, but the most common are: Staging server has DNS issues and/or required ports are blocked preventing an avtar backup issued from the staging server to succeed. Verify using the GUI you can run an on-demand backup/restore of a file from the staging server involved and that name resolution back to Avamar is successful. The utility node name listed in the environment file being used is invalid. The tape script executing on the staging server has prematurely ended, preventing the backup of the stat file. The tape script logic is broken or incomplete. The user is responsibile for ensuring that the tape policy defined to ADMe is valid and does what it is intended to do. It would be considered best practice to test this directly from the tape application as a first step to verify it behaves as intended. This policy is also responsible for the data retention on tape, therefore this too should be confirmed to be correct. When using non-incremental staging, ensure that the tape policy is configured to perform a full tape backup. With incremental tape out, you have a choice to perform full or incremental backups to tape. Auto-scheduling directly from EMC NetWorker A job policy can be initiated using a NetWorker pnpc call. When using a pnpc call, the NetWorker Management Console displays its related group as active for the duration of the migration session, and will reflect its final completion status of success or failure. However, when there is a failure, you cannot diagnose it from the NetWorker Management Console. You must use the ADMe job log to identify the cause. A NetWorker pnpc call reports on a script's return code, therefore no details are available to determine why a return code was non-zero. PNPC Prerequisites NetWorker Linux file system agent RH-64 bit must be installed on the Avamar utility node and the NetWorker service must be started using /etc/init.d/ networker start. A NetWorker client resource must be defined in the NetWorker Management Console for the utility node with its save-set field set to /etc/hosts and its backup command field set to savepnpc. Ensure its browse and retention value is set very low, such as one day or one week. This single client resource can then be leveraged by any number of NetWorker groups being used to initiate and ADMe job policy. To do so, once it has been assigned to the first NetWorker group, you can use the NetWorker Management Console to copy the group using the "copy with clients" option, renaming the new group as required. This eliminates the need to reestablish resource definitions each time with the required properties. The naming convention for the group is ADME_init_<adme-job-policy-name>. From Job Manager policy view, use the Scheduler option to automatically establish a NetWorker.res file which gets saved in /nsr/res/<ntwk-group-name>.res. Its Auto-scheduling directly from EMC NetWorker 45

46 Configuring tape applications contents will be similar to the example shown here, where the only difference will be your ADMe policy name. type: savepnpc; precmd: "/usr/local/avamar/bin/admepnpc admbatch-yourpolicy-name.cfg"; pstcmd: ""; abort precmd with group: Yes; The name of the NetWorker group must match the name of its associated NetWorker.res file being used which resides on Avamar utility node. When using the ADMe user interface to create the.res file, it will conform to the naming convention ADME_init_job-policy-name. If you choose a different naming convention, it will not display properly in the ADMe user interface. If there is more than one ADMe instance sharing the same NetWorker server, the corresponding Avamar hostnames name can be included in the group name, allowing you to distinguish which groups in NetWorker pertain to which Avamar host. This is required whenever you have common ADMe client groups on different Avamar systems. The NetWorker Schedule resource shown above is assigned to the ADMe_init_linux group and effectively determines which days and time the tape out process is to take place. The scehdule defines the days tape out is to run, and all other days must be set to skip. Once the first schedule is established it can be copied to a different name and updated as required. It is helpful to name these schedules ADME_1st, ADME_2nd, etc., so they can be easily identified and assigned to the ADME_init group as required. 46 EMC Avamar Data Migration Enabler 7.4 and Service Packs User Guide