Uploaded image for project: 'AMQ Documentation'
  1. AMQ Documentation
  2. AMQDOC-2419

HA and Failover - investigate whether we need doc for a "scale down" use case

XMLWordPrintable

    • Icon: Task Task
    • Resolution: Done
    • Icon: Major Major
    • Post-GA Update 1
    • 7.0 Beta-2
    • Broker DOC
    • None

      This was from a thread on messaging DL and sme-amq DL:

      A couple of things, If you want to scale down from a live then you would not be using live/backup pairs, you would add a scale-down configuration to th elive only policy and on clean shutdown it scales down to another live broker.

      if you want HA and scale down then it is the backup that you configure to scale down and on live failure it will start up and connect to a live broker to scale down.

      On 27 March 2017 at 21:39, Angelo Souza <asouza@redhat.com> wrote:

      Hi Dimitrii,

      The problem with this example is that the <scale-down> tag just work in the slave node and I don't know why.

      I want to scale down my master node so the messages from this master are redistributed to other master node in the cluster when i shutdown the broker

      Tks

      On Mon, Mar 27, 2017 at 5:05 PM, Dmitrii Puzikov <dpuzikov@redhat.com> wrote:

      Hi, Angelo,

      There's more than one example, please, look at this one: https://github.com/apache/activemq-artemis/blob/master/examples/features/ha/ha-policy-autobackup/src/main/resources/activemq/server0/broker.xml

      Probably, it's closer to your requirements.

      On 03/27/2017 09:04 PM, Angelo Souza wrote:

      Hi guys,

      How do we scale down a pair of Master/Slave brokers?

      We have this scenario:

      When we want to scale, we add a pair of master/slave amq7 broker, when
      scale down we remove a pair of master/slave brokers.

      The problem:

      • Documentation says /*"The simplest way to enable this behavior is
        to set |scale-down| to |true|/|/"*/ |, but where do I set this
        value to TRUE in broker.xml?
      • The examples shows other configuration, we need to add a ha-policy,
        but I'm not sure if this work in a replication ha-policy.

      <!-a live server that will scale down on server shutdown->
      <ha-policy>
      <live-only>
      <scale-down>
      <connectors>
      <connector-ref>server0-connector</connector-ref>
      </connectors>
      </scale-down>
      </live-only>
      </ha-policy>

      Is there a configuration that I can make to scale down in this scenario?

      • Scale Down a pair of master slave brokers;
      • Ha-policy = replication

      Tks,


      Ângelo Souza
      Delivery Architect | Red Hat Brasil

      M: +55 11 9 93850247

      Av. Brigadeiro Faria Lima 3900, 8° Andar. São Paulo, Brasil.
      RED HAT | TRIED. TESTED. TRUSTED. Saiba porque em redhat.com
      <https://www.redhat.com/pt-br/about/trusted><http://www.redhat.com/es/about/trusted>
      Red Hat <http://www.redhat.com.br/>


      Best Regards,
      Dmitrii Puzikov

      Quality Engineer
      Red Hat Czech s.r.o., Purkynova 647/111, 612 00 Brno, Czech Republic
      E-mail: dpuzikov@redhat.com
      phone: +420773250044
      IRC: dpuzikov at #brno, #messaging, #messaging-qe, #brno-TPB

              rhn-support-pfestoso_jira Phil Festoso (Inactive)
              sjay@redhat.com Susan Jay
              Dmitrii Puzikov Dmitrii Puzikov (Inactive)
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

                Created:
                Updated:
                Resolved: