When a bridge between 2 AMQ instances (an upstream "proxy" and a downstream "target" is configured as a static bridge and message traffic exists in the network, if the downstream broker is abnormally terminated and then restarted, the bridge is never reestablished, even if the broker is accepting client connections. Instead the upstream "proxy" broker repeatedly logs the warning:
2019-02-07 10:25:52,202 | WARN | .221:61616@52912 | DemandForwardingBridgeSupport | 139 - org.apache.activemq.activemq-osgi - 5.11.0.redhat-621222 | Network connection between vm://amq#16 and tcp://node11.dhawkins.redhat.com/10.0.0.221:61616@52912 shutdown due to a remote error: java.io.IOException: ShutdownInfo {commandId = 1, responseRequired = false}
while the downstream target broker logs:
2019-02-07 16:38:44,046 | WARN | 0.12:35198@61616 | DemandForwardingBridgeSupport | 139 - org.apache.activemq.activemq-osgi - 5.11.0.redhat-621222 | Network connection between vm://amq#84 and tcp:///10.0.0.12:35198@61616 shutdown due to a remote error: java.lang.NullPointerException
This issue is reproducible in the latest AMQ-6.2.1-621222 distribution, but does not seem to reproduce in AMQ-6.3.0-630371.