ActiveMQ broker not getting restarted in Fabric after experiencing exception from persistence store
Issue
After a Fabric managed broker received an exception from the persistence store, it is not getting restarted automatically by Fabric.
The log file stops showing more events after these lines:
INFO | ervice[mybroker] | ActiveMQServiceFactory | 101 - io.fabric8.mq.mq-fabric - 1.2.0.redhat-630224 | Found zookeeper service for broker mybroker.
INFO | ervice[mybroker] | ZooKeeperGroup | 83 - io.fabric8.fabric-groups - 1.2.0.redhat-630224 | Creating ZK Group for path "/fabric/registry/clusters/amq/default"
INFO | ervice[mybroker] | ActiveMQServiceFactory | 101 - io.fabric8.mq.mq-fabric - 1.2.0.redhat-630224 | Broker mybroker is waiting to become the master
This results in a complete broker outage!
A manual Fabric container restart is required to bring the broker back to live.
Environment
- JBoss Fuse 6.3
- R1
- R2
- R3
- R4
- Fabric mode with ActiveMQ Broker deployed to one container
- Broker received exception from persistence store (like
java.lang.IOException) and had to shut down.
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.