Forum Discussion

Vishnuaabe's avatar
Vishnuaabe
Level 4
10 years ago
Solved

Need help on resource requested for a job

Hi, 

I need your help to check is there a way to find the Logical resource is currently used somewhere ? in Netbackup 7.1.0.4.

in my shop whenever the SLPcomes to schedule they mostly queued up for a long time with the reason "Limit has been reached for the logical resource " but when I try to check them I don't find any jobs using this resource but somehow the SLPs won't run and waiting for a long for the same reason.

Is there a way I could check the availability of the resource somewhere if it is used ?.

 

 

7.1.x and Earlier
AIX
Backing Up
Backup and Recovery
NetBackup
  • Deb_Wilmot's avatar
    Deb_Wilmot
    10 years ago

    This is always a difficult issue to solve.  I would suggest first and foremost that you upgrade to 7.7 or 7.6.1.2.  The reason for the request to upgrade is because there were many, many items resolved in the later versions of NetBackup.

     

    A few of the big ones:

    1.  The 7.1 and many of the 7.5.x versions of NetBackup use a version of Sybase (NBDB/emm database) that is suspetible to random corruption of the tables.  In your case, with 7.1.0.4, a catalog recovery would have to be done to resume operations.

     

    2.  In 7.6.0.2 several long running database queries have been optimized.  The result is faster allocations.

     

    3.  In 7.6, SLP was rewritten and batching is done in a different manner.  This allows SLP's to much better. 

     

    There are a number of other issues that are resolved (along with several dataloss situations).  I would suggest that instead of troubleshooting 7.1.0.4 that you upgrade and see if the problem still occurs.

     

    Now - for what can cause the issues you're seeing (in no particular order):

     

    -  Slow disks (specifically the disks that house NBDB)

    -  Network issues as we're unable to release resouces in a timely manner under certain cirumstances.

    -  Those slow Sybase queries (#2 above).

    -  A product defect (the ones that I'm aware of are resolved in the later versions again which is again the reason for the request to upgrade).

    *** I don't speak lightly here either.  I have historically worked more resource allocation issues than any other NetBackup backline; I don't see many problems with resource allocation in the later versions, and when I do, they are almost always environment related.

     

    Please upgrade - let me know how the SLP's funciton once that occurs.

     

    Thanks,
    Deb

     

8 Replies

Sort By
  • RiaanBadenhorst's avatar
    RiaanBadenhorst
    Level 6
    10 years ago

    Post the entire message from the detailed status of the job. Part of the message would say "logical limit reached for ........." It could be the storage units stream, number of jobs on the policy, max jobs of the client etc.

  • Deb_Wilmot's avatar
    Deb_Wilmot
    Level 6
    10 years ago

    This is always a difficult issue to solve.  I would suggest first and foremost that you upgrade to 7.7 or 7.6.1.2.  The reason for the request to upgrade is because there were many, many items resolved in the later versions of NetBackup.

     

    A few of the big ones:

    1.  The 7.1 and many of the 7.5.x versions of NetBackup use a version of Sybase (NBDB/emm database) that is suspetible to random corruption of the tables.  In your case, with 7.1.0.4, a catalog recovery would have to be done to resume operations.

     

    2.  In 7.6.0.2 several long running database queries have been optimized.  The result is faster allocations.

     

    3.  In 7.6, SLP was rewritten and batching is done in a different manner.  This allows SLP's to much better. 

     

    There are a number of other issues that are resolved (along with several dataloss situations).  I would suggest that instead of troubleshooting 7.1.0.4 that you upgrade and see if the problem still occurs.

     

    Now - for what can cause the issues you're seeing (in no particular order):

     

    -  Slow disks (specifically the disks that house NBDB)

    -  Network issues as we're unable to release resouces in a timely manner under certain cirumstances.

    -  Those slow Sybase queries (#2 above).

    -  A product defect (the ones that I'm aware of are resolved in the later versions again which is again the reason for the request to upgrade).

    *** I don't speak lightly here either.  I have historically worked more resource allocation issues than any other NetBackup backline; I don't see many problems with resource allocation in the later versions, and when I do, they are almost always environment related.

     

    Please upgrade - let me know how the SLP's funciton once that occurs.

     

    Thanks,
    Deb

     

  • Vishnuaabe's avatar
    Vishnuaabe
    Level 4
    10 years ago

    Thanks Deb, I plan for an upgrade. but the still curisoity in me wants to know is there a way that we can check whether the reported resource is currently used by some other tasks.

    As Riann requested I am pasting the full messages from the job details

    08/30/2015 15:18:27 - requesting resource LCM_RDC-INF-TLD4
    08/30/2015 15:19:16 - Info nbrb (pid=61276746) Limit has been reached for the logical resource LCM_RDC-INF-TLD4

    This RDC-INF-TLD4 is noting but our storage group which only have one stroage unit that have 6 drives init.

     

  • Deb_Wilmot's avatar
    Deb_Wilmot
    Level 6
    10 years ago

    Hi Vishnuaabe,

      You might want to run:  (*** from netbackup\bin\admincmd) nbrbutil -listActiveStuJobs RDC-INF-TLD4  - this will list the currently allocated jobs for the storage unit RDC-INF-TLD4.

    Deb

     

  • Vishnuaabe's avatar
    Vishnuaabe
    Level 4
    10 years ago

    Hi All,

    here is my recent update, I have upgraded to 7.6.0.1 even after the upgrade I have the same issue.

    @Deb, even I ran this nbrbutil -listActiveStuJobs but there is nothing active for this storage unit.

    Last Thursday I opened a case with Symantec and still working on it.. will keep you posted once I get the resolutions.

    also I noticed some kind of strange activity, the SLP runs with the storage unit group if I cancell all the queued SLPs which includes the jobs uses other storage units.

     

     

  • Marianne's avatar
    Marianne
    Level 6
    10 years ago

    I would like to see output of 'nbrbutil -dump'  and 'vmoprcmd -d' on the media server when you see this.

    This RDC-INF-TLD4 is noting but our storage group which only have one stroage unit that have 6 drives init.

    If the STUG has only one STU, why use a STUG?

  • sdo's avatar
    sdo
    Moderator
    10 years ago

    The list of issues fixed between v7.6.0.1 and v7.6.0.4 is loooong... I would very much recommend a patch to v7.6.0.4 - if you are unable for sime reason to take Deb's advice of v7.6.1.2 or v7.7.x.

    Is there a reason why you stayed at v7.6.0.1 ?