You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@activemq.apache.org by "sandip.atkole" <sa...@northgateps.com> on 2018/03/26 08:40:38 UTC

Multicast discovery - Network Isolation issue

Hi,

I am using ActiveMQ version 5.15.3. I have configured ActiveMQ as Network of
brokers using multicast discovery and using KahaDB store as message
persistence.

My activemq.xml config file looks like as below:

...
<persistenceAdapter>
    <kahaDB directory="${activemq.data}/kahadb"/>
</persistenceAdapter>
...

<transportConnectors>
	<transportConnector name=&quot;openwire&quot;
uri=&quot;tcp://0.0.0.0:61616?maximumConnections=1000&amp;amp;wireFormat.maxFrameSize=104857600&quot;
discoveryUri=&quot;multicast://&lt;ip
address>:6155?group=devgrp?timeToLive=3"/> 
...
</transportConnectors>

The brokerUri/connection to the broker in client application is as below: 

"activemq:discovery:(multicast://<ip
address>:6155?group=devgrp)?transport.maxReconnectDelay=10000"

With this configuration the network of brokers/clustering works absolutely
fine where one of the brokers acts as a Primary and rest of the brokers as
Secondary. Also, when the Primary broker goes down (stops) then one of the
secondary brokers become the Primary and so on.

The issue that I am facing is when one of the brokers machine/server
isolated or goes OUT OF NETWORK then the ActiveMQ service on that server
acts as a Primary and this is as expected; but when this particular server
is back in the NETWORK the ActiveMQ service still acts as a Primary and does
not join the cluster or network of brokers. 

We would like to understand whether there is any configuration which will
helps to force the isolated ActiveMQ service to join the cluster once back
in the network.

Thanks & Regards,
Sandip



--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-Dev-f2368404.html