none
DPM 2012 - The backup job was made to fail because another Task exists for the same data source. (ID 32519) RRS feed

  • Question

  •  Hi Gents,

    We just moved from DPM 2010 to DPM 2012 on different hardware. We used the exact same settings for each protection group on the new DPM 2012 server.

    But with this new server we are getting 40-50 status failed jobs each time a backup is completed, with the following error: The backup job was made to fail because another Task exists for the same data source. (ID 32519). This error is only coming from 1 server, which has sql and file recovery point in the same protection group.

    The protection group is question has both recovery points for file & application recovery points and both start at 06:30 & 18:00.

    The errors are coming from the SQL data backup

    Any help would be appreciated

    Saturday, August 25, 2012 3:52 AM

All replies

  • Hi Billy Goat

    Do you have any maintenance jobs running on SQL during that time or that might be overlapping into that time?

    Here is a similar article:

    http://social.technet.microsoft.com/Forums/en-US/dataprotectionmanager/thread/e580a3cf-3ca6-4eb7-a9d1-3758cc45bd65/

    Saturday, August 25, 2012 7:40 AM
  •  Just checked to SQL jobs are running at either backup time.

    Nothing has changed from SQL side of things, just DPM.

    I should have mentioned that this is the same server that I am having trouble with freezing/hanging since moving to DPM 2012, which I have a question open for:

    http://social.technet.microsoft.com/Forums/en-US/dataprotectionmanager/thread/c92c17a5-9ef3-47e5-806c-5b4c2c1f61bd

    Saturday, August 25, 2012 11:24 AM
  • We have the same issue after switching from DPM 2010 to DPM 2012. In DPM 2010 we made everything looks good and green (daily status report) but now with this new issue report looks bad (see below). What is interesting is that ALL backups (Disk-To-Disk and Disk-To-Tape) are completing successfully, it just this report says that we have 150 failed jobs. Although we do not receive any alerts. To see this error I need to create custom view that shows All Failed Backup Jobs for Yesterday and then I can see these 150+ failed jobs with the same reason: The backup job was made to fail because another Task exists for the same data source. (ID 32519).

    I do not have any idea where these errors are coming from, but in My case I get these errors for all kind of sources: SQL Databases, HYPERV (VHD), Exchange DB, System State, and even for all three SharePoint farms that we have. Most of them related to SQL databases but only because number of DB on production SQL server is over 100, but the rest is all kind of sources that makes me feel that this is not specific to SQL Server but something related to DPM settings. I feel that DPM is trying to do parallel jobs for the same members (e.g. Sync and Backup), but our configuration and schedule is exactly the same as before (in DPM 2010) and we did not change anything.

    Any idea why it may be happening? If you click on more details it takes you to useless Microsoft WEB page that says that "At this time, no additional information is available for this error" :-) So far I did not see one error message that has any information available, so why they even do that? Is that supposed to be populated with some useful information? Looks like Microsoft assumes that DPM cannot have any errors...

    Anyway if anybody can suggest any solution it would be very much appreciated.

    Thanks!

    Friday, September 28, 2012 4:07 PM
  • Anybody? Any ideas?
    Monday, November 5, 2012 7:58 PM
  • I'm getting the same errors too but this is for a clustered Hyper-V child partition snapshot backup. The VHDs are stored on an HP LeftHand iSCSI SAN and have coincided with us having installed and configured the HP Application Aware Snapshot manager. I'm guessing that it's something to do with that.

    In the event log it says that there is not enough free space on the SAN even though the (thick provisioned) LUNs are less than half full. I'm guessing that this is because 98% of available space is allocated on the iSCSI SAN that holds these VHDs and it requires unallocated space to successfully create the snapshots. I'm not sure if freeing up some space will solve both problems but I'm now stuck in a Catch 22 situation where I don't want to shrink the volume size without a backup of the servers stored on them, but can't backup the servers without shrinking the volume they're stored on.

    I think the only way forward will be to move the (30+) servers to our Dell EqualLogic SAN. Re-partition the HP SAN so that there is more unallocated space avilable, then move the servers back to the HP SAN. But this will take a long amount of time so I'll not be able to report back on whether it worked or not for quite a while.

    James.

    • Edited by WorkProfile Monday, November 26, 2012 9:38 AM
    Monday, November 26, 2012 9:38 AM
  • Hi Billy,

    I had a few questions here to get a better understanding of the issue :

    1) Are there any Maintenance Plans scheduled for these SQL databases on the SQL server?

    2) When you say that Backup is not working - Please elaborate - Is the replica inconsistent or is the recovery point creation which is failing.

    3) Coudl you share the complete job failure details?

    4) Did we notice any errors on the SQL server under the event viewer on the SQL server - Check the application and system logs.

    Please provide the information requested. 

    ALSO PLEASE MAKE SURE THAT WE DO NOT HAVE MULTIPLE BACKUPS SCHEDULED FOR THE SAME SERVER IN THE SAME PROTECTION GROUP. FOR EXAMPLE - BACKUP OF C:\ D:\ SYSTEM STATE AND BMR ALONG WITH SQL DATABASES OF THE SAME SERVER UNDER 1 PROTECTION GROUP. IF THE PROTECTION GROUP IS SETUP AS LIKE THIS, THEN THE BACKUPS WILL FAIL.

    We will have to break down the protection group and stagger the backup time.


    Please remember to click “Mark as Answer” on the post that helps you. This posting is provided "AS IS" with no warranties. knowledge is valid only if it is shared by All. Regards, AJ (MSFT)


    • Edited by ANUJ_MSFT Wednesday, November 28, 2012 10:04 PM More information added
    • Proposed as answer by ANUJ_MSFT Wednesday, November 28, 2012 11:20 PM
    Wednesday, November 28, 2012 8:59 PM
  • 1) Are there any Maintenance Plans scheduled for these SQL databases on the SQL server?

    Early Responce:-

    Just checked to SQL jobs are running at either backup time.  Nothing has changed from SQL side of things, just DPM.
    2) When you say that Backup is not working - Please elaborate - Is the replica inconsistent or is the recovery point creation which is failing.

    Never mentioned backup is not working. The backups are successfully completing every time, just that we are getting errors every time the backup is completed which doesn’t look good when we get audited.

    3) Coudl you share the complete job failure details?

    OP:-
    The backup job was made to fail because another Task exists for the same data source. (ID 32519).
    4) Did we notice any errors on the SQL server under the event viewer on the SQL server - Check the application and system logs.>

    No error logs at all during the backup periods.

    The protection group also has a folder from the c:\ drive on the SQL server that it being backed up, but I don’t believe this is causing the issues because:-

    1. SQL data residing on another volume. (Not c:)
    2. There is no other backup happing on the volume the SQL data is residing on.
    3. I’m backing up 3 other SQL 2005 & 2 SQL 2008 servers in the same protection group that has a similar backup setup (i.e. backing up SQL & file data in the same protection group) without any issue.
    4. The Protection Group was setup exactly the same with DPM 2010 – was getting no error’s.

    Since the OP, I have moved the File server role from this server and I am slowly moving the SQL databases to another server. I’m looking hopefully to retire this server mid to late next year.

    In the last couple of weeks I’ve also noticed the same error’s for another server in a different protection group. This server has a SQL 2008 R2 database being backed up with DPM.  I’m not so concerned with this server though as it will be decommissioned within a week.

    Thursday, November 29, 2012 12:07 AM
  • Hi,

    So based on what I understand, the backups (synchronizations, recovery point creation and express full backups) are working just fine. We would need to see the pattern here.

    Please take a snapshot of the all jobs which had run for yesterday and today. Make sure that the failures and the success are listed. I would like to see if there is pattern here.

    Will look forward to your response.


    Please remember to click “Mark as Answer” on the post that helps you. This posting is provided "AS IS" with no warranties. knowledge is valid only if it is shared by All. Regards, AJ (MSFT)

    Thursday, November 29, 2012 3:03 AM
  • Hi,

    this seems to be related to the new feature on a protection group that allows a different schedule for file and application recovery points. If one sets those to the same schedule and the group contains both database and file backups, the problem happens.

    I have taken them apart by half an hour and it seems that the problem has gone away.

    Regards

    Uwe

    • Proposed as answer by Uwe Reisewitz Friday, March 15, 2013 2:24 PM
    Friday, March 15, 2013 2:24 PM
  • Hi,

    after 1 day I can confirm that the problem has gone away.

    Now I fail to see the purpose of those 2 schedules. It seems that the whole protection group is backed up on either schedule. Looking at the available recovery points I've seen that at least all files and all databases are backup up. In order to do this, DPM runs 2 Jobs (one for each schedule) causing the mentioned problem for the databases.

    Looks like a bug to me.

    Regards

    Uwe

    Saturday, March 16, 2013 11:34 AM
  • Yes, that exactly seems to be the issue. It looks like DPM 2012 introduced one additional backup schedule that cannot be removed or disabled, but instead can be configured to happen at least one time per week. In my case the issue was that my regular backups (Disk-to-Disk) were scheduled for 6 p.m. and also this built-in job was scheduled for 6 p.m. every day. I moved it (the built-in job) to Sunday 6 p.m. so now I have no more conflicts and my jobs are happy campers and works great.

    To access this click Protection -->

    For those who will be reading this thread and wondering where to find this new DPM 2012 built-in job please take a look at the picture below so you can check and decide when you want to make additional extra disk-to-disk backup of your servers. Happy DPM-ing!

    • Proposed as answer by lync15 Monday, March 18, 2013 1:12 PM
    Monday, March 18, 2013 1:12 PM