SOS: Missing depots from store

Good morning DP gurus, we have 3 different instances of SOS running on 3 different cell servers.  We have had several occassions, where a depot goes missing from the store but still shows up under the media pool with a status of GREEN.  This causes our copy jobs to go into mount request.  We have worked this issue extensivley with support and all they say is the medium is corrupt and can not help us.  2 of these cases, happened right afer upgrading to DP 9.05, the latest being last night.  Backup ran at 12:21am, copy job started at 5am and the depot was no longer in the store.  Thanks in advance for your help.

Parents Reply
  • did you have any media cleanup sessions running in the time period between backup and copy job start times?

    Backup start time : 12:00am

    SOS clean up job start time : 12:00am

    Copy job start time : 5:00am

     

    What's DeleteUnprotectedMediaFreq global file setting? 

    The parameter is set to twice a day:

    DeleteUnprotectedMediaFreq=2

     

Children
  • which DP version are you using exactly? Can you attach both session report for backup amd media cleanup jobs? 

  • sorry about that, we are now running DP 9.05 on linux redhot for both the CS and MA.  Please see the attached files for the backup and cleanup media jobs. - Thank you!

    Patch level         Patch description
    ===========================================
    DPUX_00128/DPLNX_00416 (A.09.05) Core Component
    DPUX_00129/DPLNX_00417 (A.09.05) Cell Manager Component
    DPUX_00130/DPLNX_00418 (A.09.05) Disk Agent
    DPUX_00131/DPLNX_00419 (A.09.05) General Media Agent
    DPUX_00132/DPLNX_00420 (A.09.05) User Interface
    DPUX_00142/DPLNX_00430 (A.09.05) StoreOnce Software Deduplication
    DPUX_00133/DPLNX_00421 (A.09.05) English Documentation (Guides, Help)
    DPUX_00128/DPLNX_00416 (A.09.05) Core Technology Stack
    DPUX_00129/DPLNX_00417 (A.09.05) Cell Server Technology Stack
    DPUX_00129/DPLNX_00417 (A.09.05) Application Server Technology Stack
    DPUX_00129/DPLNX_00417 (A.09.05) Web Services
    DPUX_00129/DPLNX_00417 (A.09.05) Java Runtime Environment Technology Stack
    DPUX_00129/DPLNX_00417 (A.09.05) Job Control Engine Service Dispatcher
    DPUX_00129/DPLNX_00417 (A.09.05) Job Control Engine Service Registry

  • 1. backup job: 

     

    [Normal] From: BMA@kbslap380.dtinet.net "KBSLAP380_StoreOnce_ARCH_gw1 [GW 12851:0:7137639611784581238]"  Time: 4/21/2016 12:00:03 AM
    	Loading medium from slot \\kbslap380.dtinet.net\ARCHIVES/6ee17499_57185043_4dbb_06ae to device KBSLAP380_StoreOnce_ARCH_gw1 [GW 12851:0:7137639611784581238]

     

    2. Media job: 

    [Normal] From: MMA@kbslap379.dtinet.net "KBSLAP379_StoreOnce_ARCH_gw2"  Time: 4/21/2016 12:14:55 AM
    	Medium: \\kbslap380.dtinet.net\ARCHIVES/6ee17499_57185043_4dbb_06ae was deleted!

     

    It's hard to tell why it was deleted without debugs from the cleanup job, but that's shouldn't be hapenning with default setting for global variable  "DeleteUnprotectedMediaMinimumAge = 86400" 

  • I just checked our global file where this issue happened and we do not have this setting in it.

  • Then it's using default value. You can set following debugging option in cell server omnirc file: 

    OB2DBG=1-500,C:102400 cleanup.txt MSM 

    Next time when used media is deleted by media cleanup provide the session reports and debugs to support. 

  • Thank you very much, at least we know what is happening now.  Not even HP support was able to provide an answer that made sense.

    Regards,

    Neiza

     

  • You might be hitting a very limited time window when media that was just created gets in the list for cleanup. You can try rescheduling your session by 5 minutes, so anything that starts at 12 AM/PM can be moved to 12:05, and then monitor if it happens again.