Scenario: We have two nodes configured in replicated colocated HA topology. Each node contains two Artemis servers, one live and one backup.
- Start both nodes
- Start sending large messages to inQueue on node 1 and receiving them from inQueue on node 1
- During sending and receiving kill node 1
- Producer and subscriber make failover on backup and continue in sending and receiving messages
- After producer sends 500 messages start node-1 again and wait for failback
Expectation: Receiver get all sent messages, none of clients gets any exception, failback was successful
Actual state: Sometimes live server on node 1 does not become active after failback.
Customer impact: After failback both live and backup are deactivated and clients cannot connect to any server.
- causes
-
JBEAP-4453 Document JBEAP-3998 for RN
- Closed
- is incorporated by
-
JBEAP-4615 Upgrade Artemis from 1.1.0.SP16 to 1.1.0.SP17
- Closed
- relates to
-
JBEAP-5245 Docs: Remove JBEAP-3998 and associated fixed JIRAs from 7.0.1 RNs
- Closed