Hello All,
I must be missing something simple here so I thought to reach out to you all.
I have just moved a clustered 2012 R2 file server from DP 7.0 to DP 9.01 and I can perform single file system backups without issues and the Shadow Copy Option is enabled by default.
This file server has 14 file systems on it and 9 needs to be backed up. Of these 9, five complete without issues but the SAME 4 have failed each day with the MEDIA agent timing out after 140 minutes (8400 seconds). If I wait for the backup to complete and try a "restart failed objects", they fail.
If I take one of these file systems and create a new backup with just that file system in it, it works OK (well, it does for the smallest file system of the four (900GB) - the others are too large to "test" during the day)
Looking in the event log on the server shows nothing of note, DP just says that it's taking a snapshot followed by the BMA aborting 140 minutes later. The original DP7 backup went to tape but this one goes to a StoreOnce 4700 Catalyst store - we see no errors related to this device.
Couple of other points
i) This server was being backed up by DP7 using Shadow Copy Services without issue including the file systems that fail.
II) We failed over the resources to another node in the Cluster that has more memory in it just in case we were running out - The same four file system failed!
I suspect this needs the complete debugs collected and sending to Support for analysis to get to the bottom of what is happening. My guess is that the disk agent is not sending data to the media agent in time but I can't work out why not.
Any ideas though before I contact Support?
Cheers
Mark