Uploaded image for project: 'Red Hat Fuse'
  1. Red Hat Fuse
  2. ENTESB-1937

patching turns example-mq into a hanged state

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Duplicate
    • Icon: Major Major
    • jboss-fuse-6.2
    • jboss-fuse-6.1 on OSE
    • Fuse on Openshift
    • None
    • % %
    • Hide
      1. install build 390-4
      2. create 3 fuse gears from OpenShift console
      3. deploy mq-amq to fuse2 (you may need to do a workaround for ENTESB-1860 )
      4. prepare patch upgrade to 390-5 by following steps from https://mojo.redhat.com/docs/DOC-983150 (except the last step)
      5. deploy the example-mq profile to fuse3
      6. do the last step in the patch upgrade process (oo-admin-upgrade upgrade-node --version <OSE version>)
      7. notice the fuse2 (the mq-amq one) and fuse3 are eventually restarted, but fuse3 never comes back to life (stays in "finalizing" state)
      Show
      install build 390-4 create 3 fuse gears from OpenShift console deploy mq-amq to fuse2 (you may need to do a workaround for ENTESB-1860 ) prepare patch upgrade to 390-5 by following steps from https://mojo.redhat.com/docs/DOC-983150 (except the last step) deploy the example-mq profile to fuse3 do the last step in the patch upgrade process (oo-admin-upgrade upgrade-node --version <OSE version>) notice the fuse2 (the mq-amq one) and fuse3 are eventually restarted, but fuse3 never comes back to life (stays in "finalizing" state)

      Attempting to patch with the testing patch (test-ose-patch-6.1.0.redhat-901.zip) breaks a running example-mq. The container appears to be hanged, only logging

      2014-09-11 07:56:44,636 | WARN  | ActiveMQ Task-3  | FailoverTransport                | sport.failover.FailoverTransport 1119 | 264 - org.apache.activemq.activemq-osgi - 5.9.0.redhat-610901 | Failed to connect to [tcp://fuse2-foobar.apps.sample.com:60313] after: 20 attempt(s) continuing to retry.
2014-09-11 08:06:44,651 | WARN  | ActiveMQ Task-3  | FailoverTransport                | sport.failover.FailoverTransport 1119 | 264 - org.apache.activemq.activemq-osgi - 5.9.0.redhat-610901 | Failed to connect to [tcp://fuse2-foobar.apps.sample.com:60313] after: 30 attempt(s) continuing to retry.

      even though the mentioned mq container ( fuse2-foobar.apps.sample.com:60313 ) appears to be alive and well.

      Note that this issue looks unrelated to https://issues.jboss.org/browse/ENTESB-1343 as in this case the container does not eventually come back to life and also the jstack suggests the condition is activemq-specific.

        1. 611r2p1_jstack.txt
          55 kB
        2. 611r2p1_karaf-example.log
          669 kB
        3. after_patch_hang_jstack.txt
          68 kB
        4. example-mq-patching.jstack.txt
          71 kB

          1.
          		 Document restart step for cartridge patching Sub-task Closed Critical Mikey Ariel

              janstey@redhat.com Jonathan Anstey
              maschmid@redhat.com Marek Schmidt
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

                Created:
                Updated:
                Resolved: