Comprehensive data protection for all workloads
Locked
Gostev
Chief Product Officer
Posts: 31599
Liked: 7092 times
Joined: Jan 01, 2006 1:01 am
Location: Baar, Switzerland
Contact:

[KNOWN ISSUES] 6.1 : Known Issues, Patches and Hotfixes

Post by Gostev »

Common issues upgrading to 6.1 from previous versions

Be sure to read and follow the upgrade procedure from the Release Notes document included with your download closely. I cannot emphasize this enough, as all upgrade issues reported so far were caused by not following the upgrade procedure.
  • You must upgrade Enterprise Manager component first, especially when it is installed on the same server with your backup management server. Otherwise, your server will get stuck in infinite reboot loop.
  • You must stop all running jobs and restores (including Instant VM Recovery sessions). Disable all near-CDP jobs temporarily to prevent them from starting during the upgrade. Failure to do saw may prevent the setup program to correctly upgrade the configuration database.
No other outstanding major issues at the moment.
Gostev
Chief Product Officer
Posts: 31599
Liked: 7092 times
Joined: Jan 01, 2006 1:01 am
Location: Baar, Switzerland
Contact:

Patch #1b (build 6.1.0.205) [September 12, 2012]

Post by Gostev »

Resolved Issues

General
• When backing up to a CIFS share, network connectivity issue during backup may cause backup storage to re-initialize, causing the following runs to fail with the “There is no FIB [summary.xml] in the specified restore point” error.
• Scheduled jobs fail to start when configuration database contains at least one record of unlicensed hypervisor host. For example, this may start happening once the licensed socket count has been exceeded.
• Transform fails with "Full point was not found" error for backup jobs that had their backup mode changed more than once.
• 1-Click FLR to the original location fails for files with ACL containing explicit ACEs and located on non-system drive.
• Failover Re-IP process fails for replica VMs with file level restore sessions active.
• Low space warning notification is not disabled when the corresponding check box is cleared.
• A number of bug fixes and performance improvements around the new GUI.

VMware
• Backup in Virtual Appliance (hot add) processing mode fails with the "File name is too long" error when VMware vCenter server has a very long name.
• Processing two VMware VMs with the same IDs (from different host) simultaneously results jobs skipping the VM with the "VM is locked by another task" event logged.
• Depending on the network topology, automatic network mode backup proxy selection algorithm may sometimes pick a proxy from a different subnet.
• Unused thick disks blocks on replica VM may contain random data instead of zeroed blocks.
• After updating ESX 4.1 to Update 3, all jobs processing VMs on that host start failing with the "Failed to parse "-quiescedsnap"" error.

Hyper-V
• On localized Hyper-V servers, restore does not work properly for VMs with SCSI disks.
• Backup of a Linux VM with Hyper-V integration tools version 3.2 or earlier fails.
• Hyper-V offhost backup fails when both VM configuration files and disks are located on the same volume.
• Hyper-V replication resets vNIC settings of the source VM, whenever replication is performed to the same host or cluster.
• Under certain environment conditions, Veeam Hyper-V Integration service may fail to start.
• Under rare circumstances, backup for Hyper-V VMs with pass-through disks may fail.

SureBackup
• SureBackup job may fail when processing domain controller VMs with the "Value was either too large or too small for an Int32" error.
• SureBackup job never stops if some VMs are present in both the application group and one of the linked jobs at the same time.
• SureBackup jobs may fail for domain controller VMs with file level restore sessions active.
• Virtual lab prevents usage of network interfaces with APIPA addresses, if VM has at least non-APIPA network interfaces.
• Virtual lab network mapping cannot be set for virtual networks with special symbols in names.

Other Enhancements

• Increased synthetic full transformation, and reverse incremental backup performance (up to a few times depending on backup storage).
• Improved Direct SAN Access mode performance by disabling excessive VDDK logging.
• Processing engine should now be much more tolerant to network packet loss.
• Added PowerShell cmdlet for VeeamZIP operation. Just as a reminder, PowerShell API is not available in Free Edition.
• Added support for application-aware processing for VMs that use EFI boot.
• Added internet access from virtual lab environment for VMs using port groups other than first.
• Increased certain network-less application aware processing timeouts to make it work more reliably with applications requiring longer time to perform VSS freeze.
• Reduced the amount of logs application-aware processing produces on guest VMs.
• Added the new registry hack that reverses the sequence of application-aware processing, making jobs try network-less processing mode before network one:

Code: Select all

InverseVssProtocolOrder
DWORD
1

B&R 6.1 Patch 1b is available for download starting September 12, 2012 (KB1671)
Gostev
Chief Product Officer
Posts: 31599
Liked: 7092 times
Joined: Jan 01, 2006 1:01 am
Location: Baar, Switzerland
Contact:

Failed to parse "-quiescedsnap"

Post by Gostev »

Issue
After updating ESX 4.1 to Update 3, all jobs processing VMs on that host start failing with the following error:

Code: Select all

Failed to parse "-quiescedsnap"
Cause
ESX 4.1 Update 3 (build 800380) introduced new functionality, which caused new unexpected "-quiescedsnap" entry to appear in VMX file (more info)

Affected versions
• All versions of Veeam B&R
• ESX 4.1 Update 3

Resolution
Patch 1 was rebuilt to include the fix (Patch 1b build 6.1.0.205), post above updated accordingly.
Locked

Who is online

Users browsing this forum: fspadaro, Gostev and 86 guests