I have a 2 node Exchange 2007 cluster running on Windows Server 2003 x64 on ESX 3.5.
Recently we shut down the filers so had to bring both nodes down, when we brought the nodes back online the SME backups started failing on the primary node with the error VSS_E_HOLD_WRITES_TIMEOUT.
When I went into SnapDrive on the primary node I noticed that the 9 RDM disks took up to 2-3 minutes to enumerate, I assumed this was causing the VSS timeout error when SME was trying to access the RDMs/Snap volumes.
I have now failed over exchange to the secondary node where SnapDrive works fine and SME is now working fine, though still fails trying to backup the passive node (the old faulty primary node) but at least we now have consistant SME snapshots!
I have tried running a re-install/repair on SnapDrive 6.0.1 (as per NetApp support recommendation) but it's mad no difference.
I'm reluctant to upgrade as you can't test untill both nodes are on the same version of SD and with only 1 stable exchange node this is seem a bit of a "all or nothing" approach!
Anyone had or seen this issue? I can't see any resolutions on NetApp/VMware Communities and NetApp support seem to be struggling.