Veritas NetBackup™ Release Notes
- About NetBackup 8.3.0.2
- New features, enhancements, and changes
- Operational notes
- About NetBackup 8.3.0.2 operational notes
- NetBackup installation and upgrade operational notes
- After initiating CA migration, connection errors may occur
- If NetBackup 8.3.0.2 upgrade fails on Windows, revert to previous log folder structure
- Native installation requirements
- NetBackup servers must use a host name that is compliant with RFC 1123 and RFC 952
- Do not install from the menu that appears when the installation DVD is inserted
- About support for HP-UX Itanium vPars SRP containers
- NetBackup administration and general operational notes
- The nbdeployutil utility may fail to generate usage reports for traditional licensing
- Child backup jobs may enter a queued state for workloads that use BigData policy
- Errors are shown in the jobs detail when NetBackup attempts to expire images from non-WORM capable storage
- NetBackup web server certificate renewal failure during initiation of NetBackup CA migration or upgrade
- Microsoft Azure backup fails if the resource group name contains a period (.)
- SLP does not retry multistreaming backup if child job fails or is canceled
- Granular restores require adequate available space on target
- Stale devices shown on the device tree
- NetBackup administration interface operational notes
- Search limitations for security events lists in the NetBackup web UI
- Using X forwarding to launch the NetBackup Administration Console can fail on certain Linux platforms
- Intermittent issues with X forwarding of NetBackup Administration Console
- NetBackup Administration Console fails in Simplified Chinese UTF-8 locale on Solaris SPARC 64-bit systems with Solaris 10 Update 2 or later
- NetBackup Bare Metal Restore operational notes
- NetBackup Cloud operational notes
- NetBackup with Veritas CloudPoint operational notes
- On-host agent may be unavailable after restore with a snapshot taken with NetBackup CloudPoint 8.3
- Application status of original instances may report errors for host-consistent restores to alternate locations
- Public cloud not supported with gov cloud or China region
- Indexing not supported on instances created from AWS Marketplaces AMIs
- Consistent host snapshot might fail
- Image clean-up may fail for Microsoft Azure workloads
- Configuring AWS plug-in with IAM role showed that the Authentication Method field is blank
- MongoDB create snapshot job may freeze
- Updating a cloud plug-in while a job runs causes job failure
- Permission denied error occurs if both user and password are updated
- Different source and target zones for Google Cloud Platform are not supported
- Broken files system detected
- NetBackup deduplication operational notes
- NetBackup for NDMP operational notes
- NetBackup internationalization and localization operational notes
- NetBackup Snapshot Client operational notes
- NetBackup virtualization operational notes
- Appendix A. About SORT for NetBackup Users
- Appendix B. NetBackup installation requirements
- Appendix C. NetBackup compatibility requirements
- Appendix D. Other NetBackup documentation and related documents
Application status of original instances may report errors for host-consistent restores to alternate locations
After restoring the host-consistent snapshot to an alternate location, the application status of the original instance that was connected using an on-host agent goes into an error state.
This issue occurs when the host-consistent snapshots that were taken on the instances with NetBackup 8.3 on-host agents are restored to an alternate location. It is recommended to follow the steps in the workaround after the restore for the host application-consistent snapshots completes.
The following error is an example of what is reported in the on-host agent logs in the source instance:
May 12 15:57:56 auto-grt-97230 flexsnap-agent-onhost[4604] MainThread
pika.channel: WARNING - Received remote Channel.Close (405):
"RESOURCE_LOCKED - cannot obtain exclusive access to locked queue
'flexsnap-agent.203f5661cdd0448bbb63138b5000e0c7' in vhost '/'.
It could be originally declared on another connection or the exclusive
property value does not match that of the original d..." on <Channel
number=1 OPEN conn=<SelectConnection OPEN socket=('10.0.0.6',
61718)->('x.x.x.x', 5671) params=<ConnectionParameters
host=x.x.x.x port=5671 virtual_host=/ ssl=True>>>
Workaround:
Follow these steps if you want to take the application-consistent snapshots of the restored instance:
On the restored host:
For a Linux on-host agent, follow the steps from the Veritas NetBackup CloudPoint Install and Upgrade Guide section Registering the Linux-based agent.
For a Windows on-host agent, follow the steps from the Veritas NetBackup CloudPoint Install and Upgrade Guide section Registering the Windows-based agent.
Note:
Ensure that you have removed the previously installed on-host agent before installation and re-registration.
On the source host:
Log in to the host.
Restart the
flexsnap-agent:Linux:
sudo systemctl stop flexsnap-agent.service
sudo systemctl start flexsnap-agent.service
Windows:
Restart Veritas NetBackup CloudPoint Agent service
Follow the steps below if you do not want to take the application-consistent snapshots of the restored instance:
On the restored host:
Follow the steps from the Veritas NetBackup CloudPoint Install and Upgrade Guide section Removing the CloudPoint agents.
Follow the steps below on the host from which the snapshot was restored:
Log in to the host.
Restart the
flexsnap-agent:Linux:
sudo systemctl stop flexsnap-agent.service
sudo systemctl start flexsnap-agent.service
Windows:
Restart Veritas NetBackup CloudPoint Agent service