Uploaded image for project: 'Lustre'
  1. Lustre
  2. LU-5266

LBUG on Failover -ldlm_process_extent_lock()) ASSERTION( lock->l_granted_mode != lock->l_req_mode )

Details

    • Bug
    • Resolution: Fixed
    • Blocker
    • Lustre 2.6.0, Lustre 2.5.4
    • Lustre 2.6.0
    • Hyperion - 2.5.60 build 2538
    • 3
    • 14694

    Description

      After hard failover of devices to server iws19, server wedged, then hit LBUG.
      Services never complete recovery, and appear to either restart the timer or something:

      Jun 27 10:08:53 iws19 kernel: Lustre: lustre-OST000c: Will be in recovery for at least 2:30, or until 316 clients reconnect
Jun 27 10:11:53 iws19 kernel: Lustre: lustre-OST000c: recovery is timed out, evict stale exports
Jun 27 10:18:30 iws19 kernel: Lustre: lustre-OST000c: Client c52d4856-d1df-b87b-911c-f1bfbc23a24d (at 192.168.124.182@o2ib) reconnecting, waiting for 316 clients in recovery for 2:27

      The server reports being cpu-bound prior to the failure
      Console log attached - unfortunately the dump after the LBUG failed.

      Attachments

        1. iws19.crash.txt
          24 kB
        2. iws23.dmesg
          29 kB
        3. iws29.lustre-log.1405540543.8118.txt
          0.3 kB
        4. iws29.lustre-log.1405540562.8181.txt
          316 kB
        5. iws29.messages.txt
          37 kB
        6. lustre-log.1404916388.10827.txt
          0.2 kB
        7. lustre-log.1404916402.10701.txt
          145 kB
        8. lustre-log.1404916421.10764.txt
          210 kB

        Issue Links

          is related to

          Bug - A problem which impairs or prevents the functions of the product. LU-5496 fix for LU-5266

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved
          is related to

          Bug - A problem which impairs or prevents the functions of the product. LU-2827 mdt_intent_fixup_resent() cannot find the proper lock in hash

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          Activity

            [LU-5266] LBUG on Failover -ldlm_process_extent_lock()) ASSERTION( lock->l_granted_mode != lock->l_req_mode )
            pjones Peter Jones added a comment -

            Vitaly
            Could you please open a new JIRA ticket to track this additional change. The original fix was in the already GA 2.6 release
            Thanks
            Peter

            pjones Peter Jones added a comment - Vitaly Could you please open a new JIRA ticket to track this additional change. The original fix was in the already GA 2.6 release Thanks Peter
            vitaly_fertman Vitaly Fertman added a comment -

            the previous fix was not exactly correct: http://review.whamcloud.com/11469

            vitaly_fertman Vitaly Fertman added a comment - the previous fix was not exactly correct: http://review.whamcloud.com/11469
            adilger Andreas Dilger added a comment -

            This should probably go into a new bug, since the LASSERT is fixed.

            adilger Andreas Dilger added a comment - This should probably go into a new bug, since the LASSERT is fixed.
            cliffw Cliff White (Inactive) added a comment -

            While testing this patch, still have some client evictions and log dumps. Message log and lustre log from latest attached.

            cliffw Cliff White (Inactive) added a comment - While testing this patch, still have some client evictions and log dumps. Message log and lustre log from latest attached.
            pjones Peter Jones added a comment -

            Landed for 2.6

            pjones Peter Jones added a comment - Landed for 2.6
            cliffw Cliff White (Inactive) added a comment -

            I tested the patch on Hyperion - no more LBUG, but did have a few evictions. lustre-logs and console logs from one run attached.

            cliffw Cliff White (Inactive) added a comment - I tested the patch on Hyperion - no more LBUG, but did have a few evictions. lustre-logs and console logs from one run attached.
            adilger Andreas Dilger added a comment -

            Was this introduced by http://review.whamcloud.com/5978 ?

            adilger Andreas Dilger added a comment - Was this introduced by http://review.whamcloud.com/5978 ?

            People

              hongchao.zhang Hongchao Zhang
              cliffw Cliff White (Inactive)
              Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: