Situation:
2 node fail over Hyper-V 2008r2 cluster, with DPM 2012 agent installed
HP Lefthand storage cluster (iSCSI), hardware snapshot software installed
10 VM's
Separate DPM 2012 Server

Problem:

The VSS writer on the Hyper-V nodes enters a failed state;

* WRITER "Microsoft Hyper-V VSS Writer"
        - Status: 10 (VSS_WS_FAILED_AT_THAW)
        - Writer failure code: 0x800423f2 (VSS_E_WRITERERROR_TIMEOUT)
        - Writer ID: {66841cd4-6ded-4f4b-8f17-fd23f8ddc3de}
        - Writer instance ID: {880a56ac-a37c-440b-9d97-b8b0346c6976}

I think this might be something to do with the back-ups within the VM's
At 21.00 a third party backup (Ahsay) is started within the VM (using VSS) on all 10 VM's.

At 23.00 the DPM back-up is started, however the back-up from 21.00 is still running.

Can this be the problem?

When I reboot my hosts to reset the VSS writer, I can take a successful backup manually (not tested at 23.00).

I provide virtual machines to my customers, they can arrange their own backup (Ahsay), however I want to back-up the VM's just to be sure, any suggestions?