Uploaded image for project: 'Lustre'
  1. Lustre
  2. LU-7754

DNE3: osd-zfs gets into a livelock if transaction is too big

Details

    • Bug
    • Resolution: Unresolved
    • Minor
    • None
    • Lustre 2.8.0
    • 3
    • 9223372036854775807

    Description

      ONLY=300k bash sanity.sh:

      [ 89.828294] LNet: Service thread pid 4249 was inactive for 40.00s. The thread might be hung, or it might only be slow and will resume later. Dumping the stack trace for debugging purposes:
      [ 89.831356] Pid: 4249, comm: mdt01_001
      [ 89.831895]
      [ 89.831895] Call Trace:
      [ 89.832451] [<ffffffff810ac6de>] ? getrawmonotonic+0x2e/0xc0
      [ 89.833222] [<ffffffff810828c5>] __cond_resched+0x25/0x40
      [ 89.834035] [<ffffffff814e521a>] _cond_resched+0x2a/0x40
      [ 89.834810] [<ffffffff814e5f11>] mutex_lock+0x11/0x40
      [ 89.835565] [<ffffffffa07741b4>] dmu_tx_assign+0x284/0x500 [zfs]
      [ 89.836410] [<ffffffffa0a1ae32>] osd_trans_start+0xb2/0x410 [osd_zfs]
      [ 89.837282] [<ffffffffa032de15>] top_trans_start+0x255/0x9c0 [ptlrpc]
      [ 89.838090] [<ffffffffa0bb88f9>] lod_trans_start+0x59/0x60 [lod]
      [ 89.838854] [<ffffffffa0ae3cdf>] mdd_trans_start+0xf/0x20 [mdd]
      [ 89.839594] [<ffffffffa0acf1a0>] mdd_create+0x1170/0x1c70 [mdd]

      Attachments

        Issue Links

          is related to

          Bug - A problem which impairs or prevents the functions of the product. LU-9096 sanity test_253: File creation failed after rm

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

          Bug - A problem which impairs or prevents the functions of the product. LU-8267 sanity test_129: (osd_internal.h:1137:osd_trans_exec_check()) LBUG

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Activity

            [LU-7754] DNE3: osd-zfs gets into a livelock if transaction is too big
            pjones Peter Jones made changes -
            Fix Version/s Original: Lustre 2.13.0 [ 14290 ]
            adilger Andreas Dilger made changes -
            Fix Version/s New: Lustre 2.13.0 [ 14290 ]
            Fix Version/s Original: Lustre 2.12.0 [ 13495 ]
            adilger Andreas Dilger made changes -
            Fix Version/s New: Lustre 2.12.0 [ 13495 ]
            pjones Peter Jones made changes -
            Fix Version/s Original: Lustre 2.11.0 [ 13091 ]
            adilger Andreas Dilger made changes -
            Summary Original: osd-zfs gets into a livelock if transaction is too big New: DNE3: osd-zfs gets into a livelock if transaction is too big
            adilger Andreas Dilger made changes -
            Labels Original: zfs New: dne3 zfs
            adilger Andreas Dilger made changes -
            Link New: This issue is related to LU-8267 [ LU-8267 ]
            pjones Peter Jones made changes -
            Fix Version/s New: Lustre 2.11.0 [ 13091 ]
            Fix Version/s Original: Lustre 2.10.0 [ 12204 ]
            adilger Andreas Dilger made changes -
            Link New: This issue is related to LU-9096 [ LU-9096 ]
            adilger Andreas Dilger added a comment -

            That is 4755 MB / 512 stripes = 9 MB/stripe which seems like a lot of space to reserve? I thought we got away from O(n^2) transaction sizes for striped directories?

            adilger Andreas Dilger added a comment - That is 4755 MB / 512 stripes = 9 MB/stripe which seems like a lot of space to reserve? I thought we got away from O(n^2) transaction sizes for striped directories?

            People

              bzzz Alex Zhuravlev
              bzzz Alex Zhuravlev
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated: