Chris Slotterback created FLINK-12448: -----------------------------------------
Summary: FlinkKafkaProducer late closure after class loader Key: FLINK-12448 URL: https://issues.apache.org/jira/browse/FLINK-12448 Project: Flink Issue Type: Bug Components: Connectors / Kafka Affects Versions: 1.7.2 Reporter: Chris Slotterback During job failure/restart, FlinkKafkaProducer configured with Semantic.EXACTLY_ONCE fails to disconnect properly do to a NoClassDefFoundError: {noformat} java.lang.NoClassDefFoundError: org/apache/kafka/clients/NetworkClient$1 at org.apache.kafka.clients.NetworkClient.processDisconnection(NetworkClient.java:658) at org.apache.kafka.clients.NetworkClient.handleDisconnections(NetworkClient.java:805) at org.apache.kafka.clients.NetworkClient.poll(NetworkClient.java:520) at org.apache.kafka.clients.producer.internals.Sender.run(Sender.java:226) at org.apache.kafka.clients.producer.internals.Sender.run(Sender.java:163) at java.lang.Thread.run(Thread.java:748){noformat} This begins a restart loop where the job never recovers properly. This is reproducible only with EXACTLY_ONCE semantic, AT_LEAST_ONCE properly disconnects and restarts without error. This issue is described in FLINK-10455, but has been since marked as Fixed, but still reproducible. -- This message was sent by Atlassian JIRA (v7.6.3#76005)