Running <APPIAN_HOME>/services/bin/status.bat (.sh) -p <password> -c in a high availability environment shows one of the Kafka brokers to be not in sync:
<APPIAN_HOME>/services/bin/status.bat (.sh) -p <password> -c
Kafka Broker Connectivity ----------------------------------------------------example1.com:9092 Reachable ISR: All replicas in sync example2.com:9092 Reachable controller ISR: Replicas not in sync not in sync: __consumer_offsets-0 not in sync: __consumer_offsets-12 not in sync: __consumer_offsets-15 not in sync: __consumer_offsets-18 not in sync: __consumer_offsets-21 not in sync: serviceManager.transaction.download-stats-0 not in sync: serviceManager.transaction.execution01-0 not in sync: serviceManager.transaction.execution02-0 not in sync: serviceManager.transaction.portal-0 example3.com:9092 Reachable ISR: All replicas in sync
Kafka Broker Connectivity ----------------------------------------------------
This issue has been addressed via AN-141918 in the following hotfixes/versions:
Apply the latest hotfix to your Appian installation or upgrade to the latest version of Appian.
It will be necessary to perform a rolling restart of the Kafka brokers on the site. The Kafka brokers that must be restarted one at a time are listed in <APPIAN_HOME>/conf/appian-topology.xml file.
<APPIAN_HOME>/conf/appian-topology.xml
<APPIAN_HOME>/services/bin/start.sh -p <password> -s zookeeper
<APPIAN_HOME>/services/bin/stop.sh -p <password> -s kafka
<APPIAN_HOME>/services/bin/start.sh -p <password> -s kafka
Repeat the above steps on each of the servers that are hosting Kafka, starting with the Replicas first, and the Controller last. The Controller can be identified from the output of the status script used in the Symptoms section.
This article applies to Appian 19.3 and earlier.