none
Two Tape Backup Fails - Data source Error ID 2019 RRS feed

  • Question

  • Windows 2012 R2
    System Center 2012 R2 DPM
    Hewlett Packard MSL G3 Series Library (x64 based)

    Issue:

    Our monthly tape backup schedule has successfully completed 437 jobs with 2 failed jobs. Both are file shares. They are schedule back up to local disk every night and this task has been successful without any issues. 

    Error message:

    The back up to tape job failed for the following reason: (ID 3311)

    View detailed errors:

    ID 2019 details: an existing connection was forcibly close by remote host (0x90072746).

    Troubleshooting taken:

    Run backup to tape job again.Restart DPMRA services.
    Restart DPM server.
    Manual RP creation straight to tape.

    Step 1.
    Please add these registry values  on DPM and Protected Server (Host) , restart DPMRA Service,  then rerun cc or manual recovery point creation job, 
    Under  "HKLM\Software\Microsoft\Microsoft Data Protection Manager\Agent"  
    Add a DWORD value with name “ConnectionNoActivityTimeoutForNonCCJobs” 
    Set it to 7200 decimal.
    Under  "HKLM\Software\Microsoft\Microsoft Data Protection Manager\Agent"
    Add a DWORD with name “ConnectionNoActivityTimeout”
    Set it to 7200 decimal.
    Step 2.
    Disable TCP Chimney on both DPM server and protected Server by running this command 
    Netsh int tcp set global chimney=disabled

    I would appreciate any help in this matter.

    Thanks.


    Sunday, September 11, 2016 9:05 PM

All replies

  • Similar problem here (after months of backing up a file share to tape without problems - suddenly this error after ~ 250 GB of ~ 750 GB total)  

     (ID 2019 Details: An existing connection was forcibly closed by the remote host (0x80072746))

    Windows Server 2012 R2 running MSSCDPM 2012 R2 version 4.2.1594.0 HP SAS LTO6 library. 

    This is the last PG which in return is part of a monthly tape backup job where everything else ran fine.

    The tape library is locally connected to the physical dpm server.  

    From the error message it appears dpm is trying to reach the source server vs just backing up from an existing disk replica. (Strange) There are plenty of recovery points available on disk. 

    Already rebooted the dpm server and tried rerunning the job as a new tape backup - same problem.

    Best regards

     Sassan

     


    Sassan Karai

    Friday, September 6, 2019 3:17 PM