Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

      Interface 1: max_sequence_id server attribute

      • Visibility: Public
      • Change Control: Stable
      • Details: Using this attribute admin can change max sequence id in server. Once sequence id reach this value server will rollback to 0 for next sequence id (i.e next job or reservation id). Default value of this attribute will be 9999999. Max value for attribute should be <= 999999999999. Min value for this attribute should be > 0.
      • Log/Error messages:
        • When non privileged user try to set max_sequence_id then following error message will be displayed by qmgr:
          • qmgr obj= svr=default: Unauthorized Request 
            qmgr: Error (15007) returned from server
        • When admin try to set invalid value for max_sequence_id then following error message will be displayed by qmgr:
          • qmgr obj= svr=default: Illegal attribute or resource value
            qmgr: Error (15014) returned from server

        • Following log message will be added to server logs when admin sets max_sequence_id:
          • Log event: PBSEVENT_ADMIN
          • Log level: LOG_NOTICE
          • Log message:svr_max_sequence_id set to val <value>
        • Following log message will be added to server logs when admin unset max_sequence_id:
          • Log event: PBSEVENT_ADMIN
          • Log level: LOG_NOTICE
          • Log message:svr_max_sequence_id reverting back to default val <value>

      Interface 2
      • Currently the largest possible Job ID is 9999999. Which is a bit low for sites submitting lots of jobs. With new enhancement the largest possible Job ID that PBS would hold now will be 999999999999.
        Thus the range will be (0 to 999999999999) with a total of 1 trillion unique job IDs. 
    • Interface 1: Qstat output
      • Visibility: Public
      • Change Control: Stable
      • Details: As PBS_MAXSEQNUM changed from 7 to 12 to support 1 trillion sequence id, so qstat output's header also got changed to adjust spaces between two field (see below space between "Job id" and "Name" field and also count of '-' in next line)
          • Previous output was:
            • Job id            Name             User              Time Use S Queue
              ----------------  ---------------- ----------------  -------- - -----
              1.vision          STDIN            pbsuser1          00:00:00 R workq 
          • New output will be:
            • Job id                 Name             User              Time Use S Queue
              ---------------------  ---------------- ----------------  -------- - -----
              1.vision               STDIN            pbsuser1          00:00:00 R workq 

    ...