Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

The host in question is a Windows Server 2008 R2 Datacenter edition host with 64GB of RAM set up as a dedicated Hyper-V server with a 2TB backing store attached via a PERC 5/e controller to a Dell MD1000. The server is a Dell Poweredge R900 Quad, quad-core.

Using Windows Server Backup it attempts to put some non-integration services available guests into saved state and use Volume Shadow Snapshots on the rest.

This takes at least 9min to do as there are at present 35 live guests, as a result the Volume Shadow Snapshot times out and the backup fails.

Is there a way to extend the timeout of Windows Server Backup to say 20 min for the volume shadow creation? Or is there a way to programatically put the guests into a saved state, run windows server backup and then bring back only the guests that were put into a saved state after the snapshot completes? This was how I back up our existing Virtual Server instances.

share|improve this question

1 Answer 1

up vote 2 down vote accepted
+250

I assume you are talking about backing up the 35 guest through the host?

If so, here are some suggestions:

  1. Split the backup into smaller sets - avoid backing all 35 guests up in a single job. This can also make recovery faster, as the backup sets are smaller.
  2. Make your backups from within each guest, rather than from the host. DR may be faster from a host-based backup, but individual file recovery is much easier when recovering from guest-based backups.
  3. If you turn off Volume Shadow Snapshots on some of the guests, will Windows Server Backup put those guests into a saved state before backing them up? If so, this may be an option.
share|improve this answer
    
Hi Brent, thanks for your answer, I do have some comments though. 1. Windows Server Backup will not let you specify individual guests to backup, it will only do volumes as far as I can tell 2. This is largely a testing environment that we provide to our internal clients, as such we cannot rely on them to have reliable backups. 3. This step also fails in the same way, times out before volume snapshot is created. This has actually started working after a power outage that exceeded our UPS runtime! Nice to know that a traumatic event fixes some things! –  Dan Nov 30 '09 at 9:31
    
I am however going to accept this answer as all the entries in it are useful and valuable to the community, I think you deserve some points for that! –  Dan Nov 30 '09 at 9:33

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.