none
Secondary DPM 2012 R2 server runs endless consistency check RRS feed

  • Question

  • Just setup new DPM 2012 R2 primary and secondary servers.  The primary is working perfect, but the secondary is running seemingly endless consistency checks on two file shares.  The check will run for hours with just a few MB transfered over a 100Mbps link.  What could be causing this?  The secondary server is backing up some SQL data from the primary just fine.


    Monday, July 7, 2014 8:15 PM

All replies

  • The initial sync has not completed yet on the two shares on the secondary server.  Sometimes the consistency check just stops like it is done, but a recovery point is never available.
    Tuesday, July 8, 2014 12:33 PM
  • Hi,

    A consistency check will not result in a recovery point - a regular recovery point job needs to succeed.  If the replica is consistent, then you can make a recovery point without a synchronization and that will make a recovery point based on the last consistency check time.

    What are details of the failed initial replica / consistency check jobs ?


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Tuesday, July 8, 2014 9:17 PM
    Moderator
  • Hi Mike,

    Thanks for the response.  One of the "consistency check" jobs has been running for 30 hours now, but only shows 158MB transfered.  Maybe it is really moving more data than it says.

    The other job has failed with the following error: (I'm going to try to kick the consistency check off again today)

    The replica of Volume C:\(protected server) on dpm server is inconsistent with the protected data source. All protection activities for data source will fail until the replica is synchronized with consistency check. You can recover data from existing recovery points, but new recovery points cannot be created until the replica is consistent. 

    For SharePoint farm, recovery points will continue getting created with the databases that are consistent. To backup inconsistent databases, run a consistency check on the farm. (ID 3106)

    An unexpected error occurred while the job was running. (ID 104 Details: The specified network name is no longer available (0x80070040))

    Thanks,

    Ryan

    Wednesday, July 9, 2014 1:11 PM
  • I tried to run another consistency check on the smaller share.  It appeared to run successfully and the check turned to green.  But, still 0 recovery points, so I though I would try to create a recovery point.  When I opened the "Create a recovery point" window my only option is to "Create a recovery point without synchronizing".  I tried to run that and receive the following error.

    Creating recovery point for C:\ failed:
    Error 208: No recovery point was created, either because synchronization has not occurred since the last recovery point was created, or because no changes were found during synchronization.
    Recommended action: If synchronization has not occurred since the last recovery point was created, you can synchronize the replica with the protected data on the protected computer and then create a recovery point. For more information, see "How to synchronize a replica" and "How to create a recovery point" in DPM Help.

    I'm guessing somehow the initial sync never ran successfully.  Should a consistency check not fix that?  What else can I try?

    Thanks.

    Wednesday, July 9, 2014 1:24 PM
  • Hi,

    Consistency checks only transfer blocks that are different between the protected servers data and the data on the replica volume.  So there maybe times where no changes are detected for hours.  Also the data transferred includes agent communications, so it's not just changed file data.

    The error (ID 104 Details: The specified network name is no longer available (0x80070040)) is usually caused by anti-virus or flaky network.


    Check Anti-virus - check exclusions.

    928840 You receive job status failure messages in Data Protection Manager
    http://support.microsoft.com/kb/928840/EN-US


    Check for Dropped packets.

    I had a customer take a network trace and saw that packets were being dropped by firewall due to packets being received in out of sync. He enabled the option for packets to get through even if they are received out of sync. This allowed backups from DPM to complete.


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Wednesday, July 9, 2014 3:45 PM
    Moderator
  • Mike,

    Should the disk allocation be going up some on the secondary server after running for that long?  For example on the primary server this replica is using ~1,660GB, but on the secondary it only shows 2.64GB used after multiple consistency checks.  The other replica is just as bad ~514GB used on the primary and only 3.42GB used on the secondary.

    I don't have anti-virus on the servers yet, but will look into possible issues with the firewall.  It is worth noting that DPM 2010 worked fine in our current WAN setup.  

    Thanks,

    Ryan

    Thursday, July 10, 2014 12:56 PM
  • how do you enable the option for packets to get through even if they are received out of sync?


    prd

    Thursday, July 10, 2014 1:12 PM
  • Mike,

    Looks like our Sonicwall IPS was blocking it!  I had to make exclusions in Sonicwall for the new servers and data is starting to flow now.  I'll confirm that this was the fix and mark your post as the answer once I get a successful replica.  Thanks for your help!

    Ryan

    Thursday, July 10, 2014 1:19 PM