Details

    • Bug
    • Resolution: Fixed
    • Major
    • Lustre 2.9.0
    • None
    • 3
    • 9223372036854775807

    Description

      Alexey Lyashkov have troubles with accessing this jira and asked my to create this issue. Here is problem description from him:

      RHEL6.5 have a backported a upstream commit 91dd8c114499e9818f2d5919ef0b9eee61810220, which broke a lustre extent handing on OFD. But Intel have a copy-paste unfixed code and use with RHEL 6.5 and up code. It's looks wrong as it have don't fixed a fiemap vs walk race while fiemap used in lustre code now (lustre commit - 144b5a65c16f1bca81ae76010528a592c821308b)
      it may produce an predictable results with memory corruption.

      attached patch backport a new api from kernel upstream already supported in lustre.

      Attachments

        Activity

          [LU-8410] fiemap vs walk race
          mdiep Minh Diep made changes -
          Link New: This issue is related to JFC-17 [ JFC-17 ]
          mdiep Minh Diep made changes -
          Link New: This issue is related to DDN-381 [ DDN-381 ]
          pjones Peter Jones made changes -
          Link New: This issue is duplicated by SEA-150 [ SEA-150 ]
          pjones Peter Jones made changes -
          Fix Version/s New: Lustre 2.9.0 [ 11891 ]
          Resolution New: Fixed [ 1 ]
          Status Original: Open [ 1 ] New: Resolved [ 5 ]
          artem_blagodarenko Artem Blagodarenko (Inactive) made changes -
          Attachment New: 0001-MRP-0000-ldiskfs-use-new-API.patch [ 22270 ]
          artem_blagodarenko Artem Blagodarenko (Inactive) created issue -

          People

            wc-triage WC Triage
            artem_blagodarenko Artem Blagodarenko (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            6 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: