locked
VM backup fails - VSS in bad state RRS feed

  • Question

  • I am using SCDPM 2012 to backup VMs on Hyper-V 2008 R2. For one VM, backup wasn't running for several weeks. Now the backup job was restarted.

    The backup job started and failed several time due to replica and recovery point threshhold exeeded. But every time I saw that the backup had transfered data. I allocated more disk space. Every time the backup job failed the Hyper-V VSS writer had a "non retryable error" when checking with vssadmin list writers. I resolved those by restarting "hyper-v virtual machine manager service". At first I thought that error was because the backup failed due to the too small recovery point/replica volumes.

    I put the volumes on autogrow. For the last backup I did not see any errors 3169 in DPM Event logs anymore. So it seems that the volume size is no longer the issue. When checking the DPMRACurr event logs it just says "AutoHeal:Not triggering CC for Apps because a CC task only marked replica inconsistent...". Because of that I am assuming that volume size is no longer the issue.

    So now I am not sure why the vss writer always shows an error message of "non retryable error". The error message in SCDPM is "VSS application writer or the VSS provider is in a bad state." The only error I saw on the hyper-v server is error ID 10112 "number of volumes reverted does not match number of volumes in the snapshot set for virtual machine". But in system/application log the snapvol/vss events are successfull/no error.

    What could I do to resolve this issue?

    Monday, February 3, 2014 10:49 PM

All replies

  • Hi

    You can also look at the MS article:

    http://support.microsoft.com/kb/975695

    • Proposed as answer by KarinExchange Friday, November 17, 2017 7:16 AM
    Friday, August 8, 2014 4:56 AM