Shashank Madhav created CAMEL-10333: ---------------------------------------
Summary: (Consumer goes into loop while processing messages) DEBUG Cannot auto-commit offsets now since the coordinator is unknown Key: CAMEL-10333 URL: https://issues.apache.org/jira/browse/CAMEL-10333 Project: Camel Issue Type: Bug Components: camel-kafka Affects Versions: 2.17.1 Environment: Spring Framework is used for creating consumers, which processes the messages. Reporter: Shashank Madhav I have a topic which have 300 messages in the queue. When the camel is used for processing the messages with event-based consumer, I am getting following error after all the messages are processed. /////error message start [ad #0 - KafkaTopic[Audit_logs]] ConsumerCoordinator DEBUG Cannot auto-commit offsets now since the coordinator is unknown, will retry after backoff [ad #0 - KafkaTopic[Audit_logs]] AbstractCoordinator DEBUG Issuing group metadata request to broker 0 [ad #0 - KafkaTopic[Audit_logs]] AbstractCoordinator DEBUG Group metadata response ClientResponse(receivedTimeMs=1473953020273, disconnected=false, request=ClientRequest(expectResponse=true, callback=org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient$RequestFutureCompletionHandler@1edc39d, request=RequestSend(header={api_key=10,api_version=0,correlation_id=9,client_id=consumer-1}, body={group_id=test3}), createdTimeMs=1473953020218, sendTimeMs=1473953020218), responseBody={error_code=0,coordinator={node_id=0,host=<host-name>,port=<port-no>}}) [ad #0 - KafkaTopic[Audit_logs]] ConsumerCoordinator ERROR Error ILLEGAL_GENERATION occurred while committing offsets for group test3 [ad #0 - KafkaTopic[Audit_logs]] ConsumerCoordinator WARN Auto offset commit failed: Commit cannot be completed due to group rebalance [ad #0 - KafkaTopic[Audit_logs]] ConsumerCoordinator ERROR Error ILLEGAL_GENERATION occurred while committing offsets for group test3 [ad #0 - KafkaTopic[Audit_logs]] ConsumerCoordinator WARN Auto offset commit failed: ////error message end Following the error, it then starts from the beginning and processes the entire queue again causing issues. The route used for camel is as follows //camel-route start <c:endpoint id="kafkaTest" uri="kafka:${broker.server}?topic=${topics1}&groupId=test3&autoOffsetReset=earliest&consumersCount=1&sessionTimeoutMs=30000&consumerRequestTimeoutMs=40000&keyDeserializer=${broker.keyDeserializer}&valueDeserializer=${broker.valueDeserializer}"/> <c:route id="route1"> <c:from uri="ref:kafkaTest"/> <c:delay> <c:constant>500</c:constant> </c:delay> <c:to uri="log:input"/> </c:route> //camel-route ends Delay is purposely added in the route to create message processing scenario. If the delay is reduced or the sessionTimeoutMs is increased then it works fine, but this is not a feasible solution. Without camel it works fine and messages are processed properly. -- This message was sent by Atlassian JIRA (v6.3.4#6332)