I have a Windows Small Business Server 2008 machine being backed up using the Windows Server Backup utility. The backups are being made to two external USB attached 1TB drives. Each drive is rotated out weekly, meaning that one drive is attached one week, then the other drive is swapped in the following week, ad infinitum for the best part of a year. Two backups were being made each day to the attached drive, one at 12pm and one at 9pm.
About two weeks ago it appears that one of the two drives reached capacity and began shedding the oldest backups, as confirmed by the VOLSNAP Event ID: 33 entries in the System Event log. On the second day, however, ALL shadow copies on the attached backup driver (volume) were deleted, as confirmed by a single VOLSNAP Event ID: 25. (See below for both Event ID message contents.) Unfortuneately, neither events were noted until the drive was swapped out with the other drive and some how the problem was propagated to the second drive, resulting in the deletion of all shadow copies on the second drive, too.
The end result... BOTH drives lost ALL previous backups leaving only the most recent full backup and no clue as to why it happened or how to prevent it from happening again. Virtually no older backup file versions could be recovered.
I was under the impression that, by design, the WSB utility would only start deleting the oldest backups (on a full drive) in order to make room for the newer backups. Granted, had I been keeping attention, I would have actually swapped out the full drives before the actually got full, so that NO data was deleted.
If I'm to continue using the Windows Server Backup utility, I need to know what happened and how to absolutely prevent it from happening again. I never expected to loose BOTH backups at the same time, and due to budget restraints, two rotating backup drives seemed a logical solution.
Here is text from the two VOLSNAP Event ID messages found in the System Log:
Event ID: 25
Source: volsnap
Level: Error
Description: The shadow copies of volume
\\?\Volume{93896cfc-d904-11dd-8cb5-001ec9ef572e} were deleted because the shadow copy storage could not grow in time. Consider reducing the IO load on the system or choose a shadow copy storage volume that is not being shadow copied.
Event ID: 33
Source: volsnap
Level: Information
Description: The oldest shadow copy of volume
\\?\Volume{93896cfc-d904-11dd-8cb5-001ec9ef572e} was deleted to keep disk space usage for shadow copies of volume
\\?\Volume{93896cfc-d904-11dd-8cb5-001ec9ef572e} below the user defined limit.
NOTE: Though I'm not sure if this has any bearing on the situation, the server is configured to make two snapshots a day of two volumes on the server, however, when I looked at the Shadow Copy settings I couldn't quite be sure if the VSS wasn't somehow also making snapshots on the two external drives, even though I haven't specified VSS to do so. None show up on the two replacement drives I'm now using, so I'd say not.
Information on how to prevent this from happening again would greatly be appreciated. (I'll be making sure no backup drive gets even close to being full from now on, that's for sure!)
Thanks,
Joseph.