I have a problem with VCB and several VMs with more than one virtual harddrives. The scripts are running fine with VMs that only have one virtual harddrive, but for VMs with more than one virtual harddrive the pre-backup script hangs and no drives are mounted. I'm running ESX 3.0.1, VC 2.0.1 and VCB 3.0.1.
Any suggestions of what I'm doing wrong?
this is a known bug within VI3 and snapshots for VCB...
The problem you are encountering is because of virtual centre naming standards call the drives identical names when they are on different LUNS, and therefore can't mount the drives because it thinks they have already been mounted... run your scripts with verbosity on the logging turned up to 6 (-L6 switch) and you'll see this error message.
You can look at renaming your vmdks using vmkfstools, but I experianced problems even after doing this.
You may want to hang on see how long the fix is going to be. or better stiil raise a call and allign it to this bug.
The other issue is that snapshots for ALL vmdks go to the home directory for the VM. This can cause some serious issues if you have high I/O on any particular VMDK. In a case like this you would end up with writes going to one LUN and reads having to hit both LUNs.
Yeah that's pretty poor, especially when you carefully placed your database VMDKs on a RAID5 volume for best performance, only for writes to be moved to your RAID1 volume when you snapshot the VM.
Gark / Kjetil,
Did you ever resolve this?
I'm having the same problem. I've moved both VMDK's to the same LUN, VCB works fine, move one of the VMDKs to a different LUN, and browse-start just hangs.... not too sure what to do next... Open an SR?
PS: The VMDK names are unique