Update 110608: The problem is caused by the fact that the volume is larger than 64TB, which is not supported in the current design of volsnap. This is documented in KB950241.
As mentioned in a previous post, I've received a STOP 0x7E dump with the following stack on Win2008R2 SP1 (a similar stack is seen on Win2008SP2):
0: kd> knL # Child-SP RetAddr Call Site00 fffff880`021dc068 fffff800`016f26ab nt!memset+0x8001 fffff880`021dc070 fffff880`01ac0c0b nt!RtlClearBits+0x8b02 fffff880`021dc0a0 fffff880`01ac50b9 volsnap!VspQueryNtfsFreeSpaceBitmap+0x2fb03 fffff880`021dc170 fffff880`01ac5ce6 volsnap!VspMarkFreeSpaceInBitmap+0x17904 fffff880`021dc360 fffff880`01ac7e21 volsnap!VspOptimizeDiffAreaFileLocation+0x2a605 fffff880`021dc6e0 fffff880`01ad769d volsnap!VspOpenDiffAreaFile+0x48106 fffff880`021dc860 fffff880`01ade5f7 volsnap!VspCreateInitialDiffAreaFile+0x1ed07 fffff880`021dc8b0 fffff880`01adf5e6 volsnap!VspTryPrepareForSnapshot+0x73708 fffff880`021dcb90 fffff880`01abd0fc volsnap!VspPrepareForSnapshot+0x11609 fffff880`021dcc50 fffff800`019c80c3 volsnap!VspPostWorker+0x2c0a fffff880`021dcc80 fffff800`016dc021 nt!IopProcessWorkItem+0x230b fffff880`021dccb0 fffff800`0196e32e nt!ExpWorkerThread+0x1110c fffff880`021dcd40 fffff800`016c3666 nt!PspSystemThreadStartup+0x5a0d fffff880`021dcd80 00000000`00000000 nt!KxStartSystemThread+0x16
VolSnap version installed is from KB2460912:
0: kd> lmtm volsnapstart end module namefffff880`01aa8000 fffff880`01af4000 volsnap Sat Dec 18 07:06:16 2010 (4D0C4F58)
I have asked one of our Developers to take a look at this. Updates will follow shortly.
Any update on the:
I have the issue on a Windows Storage Server 2008 R2 SP1 cluster protected (or not as the case is) by System Center DPM 2012.
Any news would be gratefully received. I am running:
0: kd> lmtm volsnap
start end module name
fffff880`01d37000 fffff880`01d83000 volsnap Sat Dec 18 06:06:16 2010 (4D0C4F58)
After some further investigation it seems like it crashes only when the backed up volumes are on teh same cluster node as another 80TB volume. Even though this 80TB volume isn't being backed up it's presence on the same node makes tee system crash.