Details

    • Type: Bug
    • Status: Closed
    • Priority: Major
    • Resolution: Fixed
    • Affects Version/s: 5.5.33a-galera
    • Fix Version/s: 5.5.34-galera, 5.5.34
    • Component/s: None
    • Labels:
      None
    • Environment:
      CentOS 5.x

      Description

      The Galera cluster locked up on one of the nodes. After this node was restarted, the rest of the cluster was unlocked. While it have happened with Galera, it feels more of MariaDB problem than a Galera problem.

      Log from the affected node and config is attached (other nodes don't have anything interesting in the logs and the config is exactly the same).

        Gliffy Diagrams

          Attachments

            Issue Links

              Activity

              Hide
              elenst Elena Stepanova added a comment -

              Jan,

              This is for your InnoDB expertise.

              Show
              elenst Elena Stepanova added a comment - Jan, This is for your InnoDB expertise.
              Hide
              aleksey.sanin Aleksey Sanin added a comment -

              What is the timeline for 5.5.34-galera? Or is there a patch I can apply manually? Thanks!

              Show
              aleksey.sanin Aleksey Sanin added a comment - What is the timeline for 5.5.34-galera? Or is there a patch I can apply manually? Thanks!
              Hide
              jplindst Jan Lindström added a comment -

              As soon as MariaDB 5.5.34 is released, I will merge it to MariaDB-Galera 5.5.34.

              Show
              jplindst Jan Lindström added a comment - As soon as MariaDB 5.5.34 is released, I will merge it to MariaDB-Galera 5.5.34.
              Hide
              serg Sergei Golubchik added a comment -

              reopened: started to fail after the merge with XtraDB 5.5.34

              Show
              serg Sergei Golubchik added a comment - reopened: started to fail after the merge with XtraDB 5.5.34
              Hide
              serg Sergei Golubchik added a comment -

              remerged with the latests changesets, all ok now.

              Show
              serg Sergei Golubchik added a comment - remerged with the latests changesets, all ok now.
              Hide
              jaihind213 vishnu rao added a comment -

              hi Jan,

              We hit the same error messages in production and the db froze (percona 5.5.33)
              can you provide more details on the reason for the freeze and fix ?

              our scenario was that we had very high writes (blob inserts) and high reads (blob reads) on the db.

              also could you shed some light on the message "Mutex at 0x2af53fbfc138 '&kernel_mutex'" --> what exactly is the kernel mutex ?

              thanks,
              ch Vishnu

              Show
              jaihind213 vishnu rao added a comment - hi Jan, We hit the same error messages in production and the db froze (percona 5.5.33) can you provide more details on the reason for the freeze and fix ? our scenario was that we had very high writes (blob inserts) and high reads (blob reads) on the db. also could you shed some light on the message "Mutex at 0x2af53fbfc138 '&kernel_mutex'" --> what exactly is the kernel mutex ? thanks, ch Vishnu

                People

                • Assignee:
                  jplindst Jan Lindström
                  Reporter:
                  aleksey.sanin Aleksey Sanin
                • Votes:
                  0 Vote for this issue
                  Watchers:
                  6 Start watching this issue

                  Dates

                  • Created:
                    Updated:
                    Resolved:

                    Time Tracking

                    Estimated:
                    Original Estimate - Not Specified
                    Not Specified
                    Remaining:
                    Remaining Estimate - 0 minutes
                    0m
                    Logged:
                    Time Spent - 4 hours
                    4h