FSRM causing VSS shadow copies to delete

[rayw75]

Environment:
(2) Windows 2003 R2 x64 SP2 servers in a MS Cluster as a file server.
File Server Resource manager
Volume Shadow Copy Service

Lately I've seen more frequently where my snapshots are being deleted with event ID 25:
the shadow copies of volume x were deleted because the shadow copy storage could not grow in time.....

I thought this was because the snap shot were happening during a busy time of the day but could not determine a pattern of when these are being deleted.
So I changed the snapshots to occur very early and late in the evenings. I thought this resolved the problem as the snapshots were not deleted for over 2 weeks.

Now this is happening again.

I've setup using eventtrigger.exe to e-mail me each time this event ID 25 occurs.

It turns out whenever a user hits their quota, FSRM generates a report at which time this event ID 25 occurs and the shadow copies are deleted.

I've exhausted my google-fu and cannot find any other reports FSRM causing VSS to delete shadow copies.

Has anyone else seen this?

 

[TechyMcSe2k]

VSS is the culprit that is actually deleting the older VSS data when the quota is reached. (built in by default) I would set a lower limit on the space notification so that you may have advanced warning that space is running out, or raise the quota limits.

________________________________________
Achieving a perception of high intelligence level can only be limited by your manipulation skills of the Google algorithm!

 

[rayw75]

thanks for the reply.
I would expect VSS to delete the oldest snapshots when space is running low.

However, i'm seeing all snapshots deleted when event ID 25 occurs.

Also, I have the first FSRM notification going out to me when the quota reaches %85. So the quota hasn't been reached yet.

I first get the e-mail from FSRM that %85 of the quota has been reached, then almost immediately afterwards I receive the e-mail from eventtriggers.exe that event ID 25 occurred.

 

[TechyMcSe2k]

can you post the details of the Event 25 that FSRM produces?

________________________________________
Achieving a perception of high intelligence level can only be limited by your manipulation skills of the Google algorithm!

 

[TechyMcSe2k]

There used to be a risk of overwhelming the Volume Shadow Copy Service, which causes all of your shadow copies to be deleted. It can be mitigated by placing the VSC data on a separate spindle of disks and making sure you have enough resources. This was supposed to be resolved in SP1; but if you are writing huge amounts of snapshots, then this info may help. Try moving the VSS data to another seperate LUN on another Raid Group if possible.

________________________________________
Achieving a perception of high intelligence level can only be limited by your manipulation skills of the Google algorithm!

 

[rayw75]

Stupid question:

How do I configure the VSS Storage Area to use a different drive? The drop down only has the original source drive available.

 

[TechyMcSe2k]

In the Shadow Copy Tab, you choose the storage location. When reconfiguring this you will lose all current VSS data.

http://support.microsoft.com/kb/838421

________________________________________
Achieving a perception of high intelligence level can only be limited by your manipulation skills of the Google algorithm!

 

[TechyMcSe2k]

This was supposed to be fixed in SP1...but I bet they didnt resolve 64bit issues, especially in a clustered environment.

http://support.microsoft.com/kb/826936/

Event Type: Error
Event Source: VolSnap
Event Category: None
Event ID: 25
Description:

The shadow copies of volume VolumeName were aborted because the diff area file could not grow in time. Consider reducing the IO load on this system to avoid this problem in the future.

________________________________________
Achieving a perception of high intelligence level can only be limited by your manipulation skills of the Google algorithm!