cancel
Showing results for 
Search instead for 
Did you mean: 

Unable to backup folders with DFSR running

Amanda_Jayne
Level 4

I have one set of folders on one server that I am having difficulty backing up. The server being backed up is 2008 R2, and is a "collector" for a lot of data. If I run a backup on the folder locations (selection set), I get the error 0xe000fe36 - corrupt data encountered. If I run a job on the Distributed File System for the same folders, I get the same error. The only way I can get a job to run is by stopping the DFSR service on the server, which is obviously not ideal. I am backing up other servers that have DFSR running with no problems at all.

I have set AOFO, and still no joy.

There are no issues with the VSS writers on the target server.

I have raised cases about this previously, but have not had a successful, long term resolution.

Anyone have any ideas?

BE environment is server 2003, with BE 2010 R3 with SP4, and the target server has the most up-to-date BE agent

22 REPLIES 22

marcusolini
Level 5
Employee

Hi.  Thank you for collecting the information!  I will review and discuss sometime next week.  Thank you.

marcusolini
Level 5
Employee

Hello.  Based on an initial review of the provided information, the failures for this particular job and others appear to be caused by premature and sporadic deletion of snapshots by the VSS subsystem before the backup is complete.  The premature snapshot deletion of the E: volume snapshots appear to be due to overlapping snapshots and limited Shadow Storage space on the snapshot differential area targeted to the F: volume.  For a specific example, there seems to be more than one job scheduled to back up the same volume directory E:\MH-OXA-DC-02_E.   The initial job creates a snapshot at 21/06/2014 10:04:35 and the snapshot premature deletion and failures occur at approximately 28/06/14 12:02:46.  The overlapping job creates a snapshot at 25/06/2014 00:09:27 and completes successfully at approximately 28/06/14 20:28:32.  There is also a previous job at 19/06/2014 23:13:06 that completes successfully at approximately 21/06/14 03:44:15 for the same volume directory E:\MH-OXA-DC-02_E.  In addition, automatic Shadow Copy Shared Folder snapshots are enabled for the E: volume targeted to the F: volume.  According to all of the information provided thus far, the F: volume is approximately 999 GB of which, at one point, the E: volume was using approximately 721 GB and the G: volume was using 236 GB leaving only approximately 42 GB available to share between newly created snapshots.

With all of this said…  Please review the Shadow Storage space available on the F: volume for both the E: and G: volume snapshots created by Backup Exec and Shadow Copy Shared Folders.  Also, please review the System and Application Event Logs at approximately 28/06/14 12:02:46 to confirm the premature snapshot deletion by the VSS subsystem for the specific job being investigated.  When ready and able, please let me know your thoughts.  If needed, a conference call can be arranged for a more productive discussion.  Hope this proves helpful.  Thank you.

Amanda_Jayne
Level 4

Hi marcusolini,

I've only just picked up your reply, for some reason I wasn't notified you had replied.

The problem seems to have somehow gone away by itself, possibly helped by the fact I have made sure the backup server, all storage servers and the DFSR collection server are all up to date with Windows updates and other patches. The jobs are now completing much quicker, so maybe the VSS thing isn't an issue any more.

I will montior and let you know.

Thanks for all your help.