Skip to content
JMS Bridge Docs

High Availability

Summary

High availability is a feature that allows the JMS Bridge to continue to operate even if one of the nodes in the cluster fails. This is achieved by having a standby node that can take over the work of the failed node.

Implementation is based on Artemis Shared Store Master-Slave high availability mode with important difference of shared filesystem store replaced by Kafka Streams based shared journal and Kafka Consumer Membership protocol replacing file based locks.

High Availability Architecture

JMS Bridge re-uses Shared Store activations of Artemis - but instead of using shared filesystem - Kafka Streams stateful Journal that is solely dependent on Kafka is used for Journaling operations and File based locks are replaced by Kafka Consumer Group Membership protocol - the actual flow of operations stays exactly the same.

Typical startup and failover process in Primary / Backup HA setup:

Artemis BackupHALockJournalArtemis LiveJMS ConsumerJMS ProducerArtemis BackupHALockJournalArtemis LiveJMS ConsumerJMS ProducerInitial state 'NOT_STARTED'Acquire Live LockLive Lock AcquiredAcquire Backup LockBackup Lock AcquiredAwait State != 'NOT_STARTED'Update State to 'Live'Await for State releasedAcquire Live LockRead JournalDoneStartup done, Accept connectionsSubscribeProduce Msg1Store Msg1Ack Msg1Send Msg1Ack Msg1Delete Msg1Produce Msg2Store Msg2Ack Msg2CrashLive Lock AcquiredLoad JournalDoneStartup Done, Accept ConnectionsRe-connect (Failover)Re-connect (Failover)Send Msg2AckDelete Msg2