When your backup software (eg. IBM Tivoli Data Protection for VMware) creates a quiesced snapshot on a Windows Server, sometimes one or more of the following NTFS Warnings/Errors can be found in the machines eventlogs:
- ID 50 NTFS Warning, delayed write failed / delayed write lost
- ID 57 NTFS Warning, failed to flush data to the transaction log. Courruption may occur.
- ID 137, NTFS Error, The default transaction resource manager on volume [] encountered a non-retryable error
- ID 140, NTFS Warning, failed to flush data to the transaction log. Courruption may occur in VolumeID:
- ID 12289 VSS Error, Volume Shadow Copy Service error: Unexpected error DeviceIOControl
The reason for the generated events is, that duplicate disks are created in the virtual machine.
There is a KB Article from VMware stating that this is a known Microsoft issue: Creating a quiesced snapshot of a Window 2008 R2 virtual machine generates Event IDs 50, 57, 137, 140, or 12289 (2006849).
If you do not want to ignore the errors/warnings in your eventlog you can use the following workaround:
-
open C:\ProgramData\VMware\VMware Tools\Tools.conf in a text editor (eg. notepad)
-
if the file does not exist, create it
-
add these lines to the file:
[vmbackup]
vss.disableAppQuiescing = true
-
save the file
-
exit the editor
-
restart the VMware Tools Service for the changes to take effect
Your workaround to set the parameter vss.disableAppQuiescing = true in the tools.conf is pretty dangerous because you disable VSS and after that you have no application aware snapshots anymore, only a crash consistency “backup” remains what means that you need luck that your server is in a consistency state…
You should inform your readers about that issue…
Bye Tom
yes I agree on that, as this gives you less of a guarantee to be able to recover your system doing a restore.
We will open an incident @ IBM on this subject.