Uploaded image for project: 'JBoss Enterprise Application Platform'
  1. JBoss Enterprise Application Platform
  2. JBEAP-1527

Clients wait for connection timeout during failback

    XMLWordPrintable

Details

    • Bug
    • Resolution: Done
    • Major
    • 7.0.0.ER6
    • 7.0.0.DR11
    • ActiveMQ
    • None
    • Hide
      • Start two nodes Node1 (live) Node2 (backup)
      • Start clients sending and receiving messages on Node1
      • Kill Node 1, clients execute failover sequence to backup Node2.
      • Continue sending and receiving messages
      • Start Node1 again and wait for clients to failback.
      Show
      Start two nodes Node1 (live) Node2 (backup) Start clients sending and receiving messages on Node1 Kill Node 1, clients execute failover sequence to backup Node2. Continue sending and receiving messages Start Node1 again and wait for clients to failback.

    Description

      Clients failback is slow on HA topology with shared journal. According to log information clients wait until connections timeout. I encountered this wit AUTO_ACK and also TRANS_ACK clients. Logs are in attachment.

      Attachments

        1. failback_autoack
          90 kB
        2. failback_transack
          69 kB
        3. server-trace1.log
          3.88 MB
        4. server-trace2.log
          1.69 MB
        5. server-trace3.log
          904 kB

        Issue Links

          Activity

            People

              thofman Tomas Hofman
              okalman@redhat.com Ondřej Kalman (Inactive)
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: