Uploaded image for project: 'AMQ Interconnect'
  1. AMQ Interconnect
  2. ENTMQIC-1992

Enabling AMQ 7.1 broker failback breaks failover support

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Done
    • Priority: Major
    • Resolution: Done
    • Affects Version/s: 1.1.0.GA
    • Fix Version/s: 1.2.0.CR2
    • Component/s: None
    • Labels:
      None
    • Environment:

      Red Hat Enterprise Linux Server release 7.2 (Maipo)

    • Target Release:
    • Sprint:
      Interconnect - June Sprint
    • Steps to Reproduce:
      Hide

      See attached configuration.

      • Start broker-master
      • Once broker-master is up start broker-slave
      • Start the router
      • Test sending and receiving and failoverList
      • Stop broker-master, backup slave should become live
      • Test sending and receiving and failoverList
      • Restart broker-master and allow it to become live and slave backup once again
      • Test sending and receiving and failoverList
      Show
      See attached configuration. Start broker-master Once broker-master is up start broker-slave Start the router Test sending and receiving and failoverList Stop broker-master, backup slave should become live Test sending and receiving and failoverList Restart broker-master and allow it to become live and slave backup once again Test sending and receiving and failoverList
    • Affects:
      Release Notes
    • Release Notes Text:
      Hide
      Previously, if a router was connected to the master instance of a broker cluster, and then failed over to the backup instance, it was possible for the router to lose track of the master instance and fail to reconnect to it once it was back online. This issue has been fixed to ensure that a router can always reconnect to the original master instance in a broker cluster.
      Show
      Previously, if a router was connected to the master instance of a broker cluster, and then failed over to the backup instance, it was possible for the router to lose track of the master instance and fail to reconnect to it once it was back online. This issue has been fixed to ensure that a router can always reconnect to the original master instance in a broker cluster.
    • Release Notes Docs Status:
      Documented as Resolved Issue

      Description

      This may be a broker related problem, however from the end developer's perspective...

      Upon configuring two brokers for failover and a single router as per ENTMQIC-1975 I see that the router correctly fails over to the slave and sending messages is successful both before and after the single failover.
      Initial failover data:

      [0x2534010]:0 <- @open(16) [container-id="broker-master", max-frame-size=4294967295, channel-max=65535, idle-time-out=30000, offered-capabilities=@PN_SYMBOL[:"sole-connection-for-container", :"DELAYED_DELIVERY", :"SHARED-SUBS", :"ANONYMOUS
      		 
      -RELAY"], properties={:product="apache-activemq-artemis", :"failover-server-list"=[{:hostname="192.168.2.208", :scheme="amqp", :port=62616, :"network-host"="192.168.2.208"}], :version="2.4.0.amq-710004-redhat-1"}]

      An an initial qdmanage query of the failoverList

      [rkieley@ic1rh configuration]$ qdmanage -b amqp://192.168.2.208:5672 --type=connector query failoverList                                                                                                                                       
      [
      		 
        {
      		 
          "failoverList": "amqp://192.168.2.208:5672"
      		 
        }
      		 
      ]

      Both prior to failover and after I can send and receive messages via the live broker successfully.

      However when failover occurs I do not see any failover related information being received by the router and a subsequent check of the failoverList reveals:

      [rkieley@ic1rh configuration]$ qdmanage -b amqp://192.168.2.208:5672 --type=connector query failoverList
      		 
      []
      		 
      [rkieley@ic1rh configuration]$

      Once failback occurs I can no longer successfully send messages via the waypointed address.

        Gliffy Diagrams

          Attachments

            Issue Links

            is related to

            Epic - Created by Jira Software - do not edit or delete. Issue type for a big user story that needs to be broken down. ENTMQIC-1920 Honor alternate URL provided by connected brokers

            • Major - A request that should be considered seriously but is not a show stopper.
            • Done

            Bug - A problem which impairs or prevents the functions of the product. ENTMQIC-1975 failover not working with master/slave Artemis brokers

            • Blocker - An issue (bug, feature, task) that blocks development and/or testing work, production could not run. An upcoming version that is affected by this issue cannot be released until it's addressed.
            • Reopened

            Bug - A problem which impairs or prevents the functions of the product. ENTMQIC-1994 When using linkRoute and AMQP client failover, connection is not reestablished if broker connection is lost

            • Major - A request that should be considered seriously but is not a show stopper.
            • New
            relates to

            Bug - A problem which impairs or prevents the functions of the product. ENTMQBR-1763 Acceptor is available before HA replication with backup server is complete, causing fail-over-list to be empty

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

              Activity

                People

                • Assignee:
                  ganeshmurthy Ganesh Murthy
                  Reporter:
                  roddiekieley Roddie Kieley
                • Votes:
                  0 Vote for this issue
                  Watchers:
                  6 Start watching this issue

                  Dates

                  • Created:
                    Updated:
                    Resolved: