Veritas NetBackup™ Deduplication Guide
- Introducing the NetBackup media server deduplication option
- Planning your deployment
- Planning your MSDP deployment
- NetBackup naming conventions
- About MSDP deduplication nodes
- About the NetBackup deduplication destinations
- About MSDP storage capacity
- About MSDP storage and connectivity requirements
- About NetBackup media server deduplication
- About NetBackup Client Direct deduplication
- About MSDP remote office client deduplication
- About the NetBackup Deduplication Engine credentials
- About the network interface for MSDP
- About MSDP port usage
- About MSDP optimized synthetic backups
- About MSDP and SAN Client
- About MSDP optimized duplication and replication
- About MSDP performance
- About MSDP stream handlers
- MSDP deployment best practices
- Use fully qualified domain names
- About scaling MSDP
- Send initial full backups to the storage server
- Increase the number of MSDP jobs gradually
- Introduce MSDP load balancing servers gradually
- Implement MSDP client deduplication gradually
- Use MSDP compression and encryption
- About the optimal number of backup streams for MSDP
- About storage unit groups for MSDP
- About protecting the MSDP data
- Save the MSDP storage server configuration
- Plan for disk write caching
- Provisioning the storage
- Licensing deduplication
- Configuring deduplication
- Configuring MSDP server-side deduplication
- Configuring MSDP client-side deduplication
- About the MSDP Deduplication Multi-Threaded Agent
- Configuring the Deduplication Multi-Threaded Agent behavior
- Configuring deduplication plug-in interaction with the Multi-Threaded Agent
- About MSDP fingerprinting
- About the MSDP fingerprint cache
- Configuring the MSDP fingerprint cache behavior
- About seeding the MSDP fingerprint cache for remote client deduplication
- Configuring MSDP fingerprint cache seeding on the client
- Configuring MSDP fingerprint cache seeding on the storage server
- Enabling 96-TB support for MSDP
- Configuring a storage server for a Media Server Deduplication Pool
- Configuring a storage server for a PureDisk Deduplication Pool
- About disk pools for NetBackup deduplication
- Configuring a disk pool for deduplication
- Creating the data directories for 96-TB MSDP support
- Adding volumes to a 96-TB Media Server Deduplication Pool
- Configuring a Media Server Deduplication Pool storage unit
- Configuring client attributes for MSDP client-side deduplication
- Disabling MSDP client-side deduplication for a client
- About MSDP compression
- About MSDP encryption
- MSDP compression and encryption settings matrix
- Configuring encryption for MSDP backups
- Configuring encryption for MSDP optimized duplication and replication
- About the rolling data conversion mechanism for MSDP
- Modes of rolling data conversion
- MSDP encryption behavior and compatibilities
- Configuring optimized synthetic backups for MSDP
- About a separate network path for MSDP duplication and replication
- Configuring a separate network path for MSDP duplication and replication
- About MSDP optimized duplication within the same domain
- Configuring MSDP optimized duplication within the same NetBackup domain
- About MSDP replication to a different domain
- Configuring MSDP replication to a different NetBackup domain
- About configuring MSDP optimized duplication and replication bandwidth
- About storage lifecycle policies
- About the storage lifecycle policies required for Auto Image Replication
- Creating a storage lifecycle policy
- About MSDP backup policy configuration
- Creating a backup policy
- Resilient Network properties
- Specifying resilient connections
- Adding an MSDP load balancing server
- About the MSDP pd.conf configuration file
- Editing the MSDP pd.conf file
- About the MSDP contentrouter.cfg file
- About saving the MSDP storage server configuration
- Saving the MSDP storage server configuration
- Editing an MSDP storage server configuration file
- Setting the MSDP storage server configuration
- About the MSDP host configuration file
- Deleting an MSDP host configuration file
- Resetting the MSDP registry
- About protecting the MSDP catalog
- Changing the MSDP shadow catalog path
- Changing the MSDP shadow catalog schedule
- Changing the number of MSDP catalog shadow copies
- Configuring an MSDP catalog backup
- Updating an MSDP catalog backup policy
- Configuring deduplication to the cloud with NetBackup CloudCatalyst
- Using NetBackup CloudCatalyst to upload deduplicated data to the cloud
- CloudCatalyst requirements and limitations
- Configuring a Linux media server as a CloudCatalyst storage server
- Configuring a CloudCatalyst storage server for deduplication to the cloud
- About the CloudCatalyst esfs.json configuration file
- About the CloudCatalyst cache
- Controlling data traffic to the cloud when using CloudCatalyst
- Configuring push or pull optimized duplication for CloudCatalyst
- Decommissioning CloudCatalyst cloud storage
- NetBackup CloudCatalyst workflow processes
- Disaster Recovery for CloudCatalyst
- Monitoring deduplication activity
- Managing deduplication
- Managing MSDP servers
- Viewing MSDP storage servers
- Determining the MSDP storage server state
- Viewing MSDP storage server attributes
- Setting MSDP storage server attributes
- Changing MSDP storage server properties
- Clearing MSDP storage server attributes
- About changing the MSDP storage server name or storage path
- Changing the MSDP storage server name or storage path
- Removing an MSDP load balancing server
- Deleting an MSDP storage server
- Deleting the MSDP storage server configuration
- Managing NetBackup Deduplication Engine credentials
- Managing Media Server Deduplication Pools
- Viewing Media Server Deduplication Pools
- Determining the Media Server Deduplication Pool state
- Changing Media Server Deduplication Pool state
- Viewing Media Server Deduplication Pool attributes
- Setting a Media Server Deduplication Pool attribute
- Changing a Media Server Deduplication Pool properties
- Clearing a Media Server Deduplication Pool attribute
- Determining the MSDP disk volume state
- Changing the MSDP disk volume state
- Inventorying a NetBackup disk pool
- Deleting a Media Server Deduplication Pool
- Deleting backup images
- About MSDP queue processing
- Processing the MSDP transaction queue manually
- About MSDP data integrity checking
- Configuring MSDP data integrity checking behavior
- About managing MSDP storage read performance
- About MSDP storage rebasing
- About the MSDP data removal process
- Resizing the MSDP storage partition
- How MSDP restores work
- Configuring MSDP restores directly to a client
- About restoring files at a remote site
- About restoring from a backup at a target master domain
- Specifying the restore server
- Managing MSDP servers
- Recovering MSDP
- Replacing MSDP hosts
- Uninstalling MSDP
- Deduplication architecture
- Troubleshooting
- About unified logging
- About legacy logging
- NetBackup MSDP log files
- Troubleshooting MSDP installation issues
- Troubleshooting MSDP configuration issues
- Troubleshooting MSDP operational issues
- Verify that the MSDP server has sufficient memory
- MSDP backup or duplication job fails
- MSDP client deduplication fails
- MSDP volume state changes to DOWN when volume is unmounted
- MSDP errors, delayed response, hangs
- Cannot delete an MSDP disk pool
- MSDP media open error (83)
- MSDP media write error (84)
- MSDP no images successfully processed (191)
- MSDP storage full conditions
- Troubleshooting MSDP catalog backup
- Viewing MSDP disk errors and events
- MSDP event codes and messages
- Troubleshooting CloudCatalyst issues
- CloudCatalyst logs
- Problems encountered while using the Cloud Storage Server Configuration Wizard
- Disk pool problems
- Problems during cloud storage server configuration
- Status 191: No images were successfully processed
- Media write error (84) if due to a full local cache directory
- Restarting the vxesfsd process
- Problems restarting vxesfsd
- CloudCatalyst troubleshooting tools
- Appendix A. Migrating to MSDP storage
About seeding the MSDP fingerprint cache for remote client deduplication
Veritas provides a method for seeding the fingerprint cache for a new client. The use case that benefits the most from seeding is the first backup of a remote client over a high latency network such as a WAN. The performance of the first backup is then similar to the performance of an existing client.
An important consideration is the client from which to seed the cache. When you choose a similar client, consider the following:
If most of the information is the operating system files, use any client with the same operating system.
If most of the information is data, finding a client with the same data may be unlikely. Therefore, consider physically moving a copy of the data to the datacenter. Back up that data on a similar client, and then use that client and policy for the seed.
The more similar the clients are, the greater the cache hit rate is.
Two methods exist to configure cache seeding. You can use either method. The following table describes the seeding configuration methods.
Table: Seeding configuration methods
Host on which to configure seeding | Description |
|---|---|
On the client | Configure seeding on the client for one or only a few clients. See Configuring MSDP fingerprint cache seeding on the client. |
On the storage server | The use case that benefits the most is many clients to seed, and they can use the fingerprint cache from a single host. See Configuring MSDP fingerprint cache seeding on the storage server. |
To ensure that NetBackup uses the seeded backup images, the first backup of a client after you configure seeding must be a full backup with a single stream. Specifically, the following two conditions must be met in the backup policy:
The Attributes tab attribute must be unchecked.
The backup selection cannot include any directives.
If these two conditions are not met, NetBackup may use multiple streams. If the Attributes tab is set to a number less than the total number of streams, only those streams use the seeded images to populate the cache. Any streams that are greater than the value do not benefit from seeding, and their cache hit rates may be close to 0%.
After the first backup, you can restore the original backup policy parameter settings.
The following items are example of informational messages that show that seeding occurred:
Activity Monitor Job Details | 1/2/2015 2:18:23 AM - Info nbmaster1(pid=6340) StorageServer=PureDisk:nbmaster1; Report=PDDO Stats for (nbmaster1): scanned: 3762443 KB, CR sent: 1022 KB, CR sent over FC: 0 KB, dedup: 100.0%, cache hits: 34364 (100.0%) 1/2/2015 2:18:24 AM - Info nbmaster1(pid=6340) StorageServer=PureDisk:nbmaster1; Report=PDDO Stats for (nbmaster1): scanned: 1 KB, CR sent: 0 KB, CR sent over FC: 0 KB, dedup: 100.0% |
Deduplication plug-in log ( | 01/02/15 02:15:17 [4452] [4884] [DEBUG] PDSTS: cache_util_get_cache_dir: enter db=/nbmaster1#1/2, scp='', bc=opscenter1, bp=seedfinal, bl=4096 01/02/15 02:15:17 [4452] [4884] [DEBUG] PDSTS: cache_util_get_cache_dir: new backup, using existing client seeding directory 01/02/15 02:15:17 [4452] [4884] [DEBUG] PDSTS: cache_util_get_cache_dir: exit db=/nbmaster1#1/2, scp='', bc=opscenter1, bp=seedfinal, bl=4096, cachedir_buf='/nbmaster1#1/2/#pdseed/opscenter1' err=0 |
Deduplication proxy server log ( | 02:15:17.417[4452.4884][DEBUG][dummy][11:bptm:6340:nbmaster1][DEBUG] PDSTS: cache_util_get_cache_dir: enter db=/nbmaster1#1/2, scp='', bc=opscenter1, bp=seedfinal, bl=4096 02:15:17.433[4452.4884][DEBUG][dummy][11:bptm:6340:nbmaster1][DEBUG] PDSTS: cache_util_load_fp_cache_nbu: enter dir_path=/nbmaster1#1/2/#pdseed/opscenter1, t=16s, me=1024 02:15:17.449[4452.4884][DEBUG][dummy][11:bptm:6340:nbmaster1][DEBUG] PDSTS: cache_util_load_fp_cache_nbu: adding 'nbmaster1_1420181254_C1_F1.img' to cache list (1) 02:15:17.449[4452.4884][DEBUG][dummy][11:bptm:6340:nbmaster1][DEBUG] PDSTS: cache_util_load_fp_cache_nbu: opening /nbmaster1#1/2/#pdseed/opscenter1/nbmaster1_1420181254_C1_F1.img for image cache (1/1) 02:15:29.585[4452.4884][DEBUG][dummy][11:bptm:6340:nbmaster1][DEBUG] PDVFS: pdvfs_lib_log: soRead: segment c32b0756d491871c45c71f811fbd73af already present in cache. 02:15:29.601[4452.4884][DEBUG][dummy][11:bptm:6340:nbmaster1][DEBUG] PDVFS: pdvfs_lib_log: soRead: segment 346596a699bd5f0ba5389d4335bc7429 already present in cache. |
For more information about seeding, see the following Veritas tech note: