Forum Discussion

Shas's avatar
Shas
Level 4
11 years ago
Solved

Media server bounce back to Active for disk in SSO

Hi All,

We have a newly setup media servers which is showing Active for Disk only instead of Active for Disk & Tape . SSO licence is active on the Media server

While configuration it shows the drive path and drives are visible from the command "vmoprcmd -d " for few seconds and then the drives go invisible as given below :

D:\Veritas\Volmgr\bin>vmoprcmd -d


                                PENDING REQUESTS

                                     <NONE>

                                  DRIVE STATUS

Drv Type Control User Label RecMID ExtMID Ready Wr.Enbl. ReqId

                             ADDITIONAL DRIVE STATUS

Drv DriveName Shared Assigned Comment

D:\Veritas\Volmgr\bin>

 

We tried to force it from the command , but no use :

nbemmcmd -updatehost -machinename client_name -machinestateop set_tape_active -machinetype media -masterserver <Master server>

NBU Version: 7.5.0.6

OS : win2008 R2

I am attaching the LTID debug log.

  • There are connectivity errors.

    Media servers can go offline for tape (active for disk only) for many reasons including the following:

       - Master failed to receive the hearbeats.

        -  Drive polling fails.

    If ltid is exiting it's because it can't detect any devices on the system. 

     

    In this case ltid starts up, but detects a change in machine state:
    17:36:15.121 [904.1212] <4> avrd: INITIATING
    17:36:15.121 [6968.4420] <4> SendEmmHeartbeat: Detected change in MachineState...
    17:36:15.121 [6968.4420] <4> SendEmmHeartbeat: ...clearing LOCAL_CONTROL bits
     

    So apparently the heartbeats failed.

    Every 5 minutes the media server will attempt to send the heart beat again...  Every time after 17:36 - it Detects a change in machine state and tries to clear local bits and up the libraries.

     

    At this point all updates fail:

    18:06:15.442 [6968.4420] <16> emmlib_SetRobotStatus: (0) UpdateLibrary failed, emmError = 2007031, nbError = 0
    18:06:15.442 [6968.4420] <16> ROBOT_CONNECT: (-) Translating EMM_ERROR_SQLNoDataFound(2007031) to 328 in the device management context
    18:06:15.442 [6968.4420] <16> ROBOT_CONNECT: emmlib_SetRobotStatus() failed for robot 3 with error 328
    18:06:15.442 [6968.4420] <4> SendEmmHeartbeat: Going Active failed - reset machine state.
     

    I would suggest you check the following:

    WINS or DNS server issues

    A NIC going bad.

    You might want to schedule a bptestnetconn every 10 minutes or so to see when the media sever is not available.

    I woul suggest enabling the folloiwng logigng levels on the media server:

    vxlogcfg -a -p 51216 -o 111 -s DebugLevel=6 -s DiagnosticLevel=6

    vxlogcfg -a -p 51216 -o 156 -s DebugLevel=6 -s DiagnosticLevel=6

     

    Make sure that the daemon directory is also available under volmgr\debug

    On the maser - if space allows - run the following:
     

    vxlogcfg -a -p 51216 -o 111 -s DebugLevel=6 -s DiagnosticLevel=6

    vxlogcfg -a -p 51216 -o 156 -s DebugLevel=6 -s DiagnosticLevel=6

    vxlogcfg -a -p 51216 -o 137-s DebugLevel=6 -s DiagnosticLevel=6

     

    Check the resulting log files for issues connecting.

    An example is in http://www.symantec.com/docs/TECH69625

     

    *** Don't forget to remove logging when done.

    Media server:

    vxlogcfg -r -p 51216 -o 111 -s DebugLevel=6 -s DiagnosticLevel=6

    vxlogcfg -r -p 51216 -o 156 -s DebugLevel=6 -s DiagnosticLevel=6

     

    Master:
     

    vxlogcfg -r -p 51216 -o 111 -s DebugLevel=6 -s DiagnosticLevel=6

    vxlogcfg -r -p 51216 -o 156 -s DebugLevel=6 -s DiagnosticLevel=6

    vxlogcfg -r -p 51216 -o 137-s DebugLevel=6 -s DiagnosticLevel=6

     

    Good luck

     

     

     

     

     

    After that happens this entity can no longer make any updates to the database and