OPERATIONAL DEFECT DATABASE
...
...
After backing up a virtual machine on an ESX/ESXi 3.5 host using a third party backup software, such as Symantec Backup Exec, you experience these symptoms:Backup completes successfully, but snapshots are not removed from the virtual machineThe virtual machine's snapshot manager shows the snapshot from the backup softwareThe files are not locked and you are able to remove the snapshots using snapshot managerIn the backup software, you see that the backup started and completed successfully:Taking backup[5888] 10/12/11 03:42:07 [ndmp\loops] - LP_ENV::createVMWareSnapshot(): creating a snapshot 'SYMC-FULL 12-10-2011 03:42' for the vm '(DC)99 Sum(DC)\vm\Production\Win\notesmailsvr'[5888] 10/12/11 03:42:07 [fsys\shared] - vddkCreateVMWareSnapshot() creating a snapshot 'SYMC-FULL 12-10-2011 03:42' for the VM '(DC)99 Sum(DC)\vm\Production\Win\notesmailsvr'Removing backup[5888] 10/12/11 04:06:58 [ndmp\loops] - LP_ENV::removeVMWareSnapshot(): removing the snapshot 'SYMC-FULL 12-10-2011 03:42' for the VM '(DC)99 Sum(DC)\vm\Production\Win\notesmailsvr'[5888] 10/12/11 04:06:58 [fsys\shared] - vddkRemoveSnapshot(): removing the snapshot 'SYMC-FULL 12-10-2011 03:42' for the VM '(DC)99 Sum(DC)\vm\Production\Win\notesmailsvr'In the virtual machine vmware.log log files, you see the create snasphot event, but not the corresponding request to delete the snapshot:Oct 12 03:42:26.682: vmx| SnapshotVMX_TakeSnapshot start: 'SYMC-FULL 12-10-2011 03:42', powerState=0, deviceState=0, logging=0, cb=8254CB0, cbData=8859B90Oct 12 03:42:26.818: vmx| DUMPER: Creating checkpoint file /vmfs/volumes/494c447a-ee3d507c-9bdf-00215e2ef162/notesmailsvr/notesmailsvr-Snapshot3299.vmsnOct 12 03:42:27.533: vmx| SnapshotVMX done with snapshot 'SYMC-FULL 12-10-2011 03:42': 3299 In the /var/log/messages log file, you see entries similar to:Oct 12 04:06:59 lvlesx1 VMware[init]: [1]+ Aborted (core dumped) setsid $CMDOct 12 04:06:59 lvlesx1 watchdog-hostd: '/usr/sbin/vmware-hostd -u' exited after 1866 secondsOct 12 04:06:59 lvlesx1 watchdog-hostd: Executing cleanup command '/usr/sbin/vmware-hostd-support'Oct 12 04:07:01 lvlesx1 watchdog-vpxa: '/opt/vmware/vpxa/sbin/vpxa' exited after 1870 secondsThe /var/log/hostd log showing the following message and the hostd process restarts:[2011-10-12 03:35:43.809 'Vmsvc' 56138672 panic] Failed to clone a context from foundry
This is a known issue with ESX/ESXi 3.5.To resolve this issue, upgrade the ESX/ESXi 3.5 host to ESX 3.5 patch 23. For more information, see:ESX 3.5 – VMware ESX 3.5, Patch ESX350-201008401-SG: Updates VMkernel, hostd, and VMX (1026126)ESXi 3.5 – VMware ESXi 3.5, Patch ESXe350-201008401-I-SG: Updates firmware (1026138)
Click on a version to see all relevant bugs
VMware Integration
Learn more about where this data comes from
Bug Scrub Advisor
Streamline upgrades with automated vendor bug scrubs
BugZero Enterprise
Wish you caught this bug sooner? Get proactive today.
