No binlogs after connection retry in maxwell

[manishsingh-rzp]

I got a warn in maxwell pod runnin on kubernetes .

18/10/2024 22:17:23.439 2024-10-18 16:47:23 WARN BinlogConnectorReplicator - communications failure in binlog: 18/10/2024 22:17:23.439 com.github.shyiko.mysql.binlog.network.ServerException: The last transaction was aborted due to Zero Downtime Restart. Please retry. 18/10/2024 22:17:23.439 \tat com.github.shyiko.mysql.binlog.BinaryLogClient.listenForEventPackets(BinaryLogClient.java:1043) [mysql-binlog-connector-java-0.27.4.jar:0.27.4] 18/10/2024 22:17:23.439 \tat com.github.shyiko.mysql.binlog.BinaryLogClient.connect(BinaryLogClient.java:631) [mysql-binlog-connector-java-0.27.4.jar:0.27.4] 18/10/2024 22:17:23.439 \tat com.github.shyiko.mysql.binlog.BinaryLogClient$7.run(BinaryLogClient.java:932) [mysql-binlog-connector-java-0.27.4.jar:0.27.4] 18/10/2024 22:17:23.439 \tat java.lang.Thread.run(Thread.java:829) [?:?] 18/10/2024 22:17:23.439 2024-10-18 16:47:23 INFO BinlogConnectorReplicator - Binlog disconnected. 18/10/2024 22:17:23.537 2024-10-18 16:47:23 WARN BinlogConnectorReplicator - replicator stopped at position: mysql-bin-changelog.009885:97551300 -- restarting 18/10/2024 22:17:23.537 2024-10-18 16:47:23 INFO BinlogConnectorReplicator - Reconnection attempt: 1 of 1 18/10/2024 22:17:23.554 2024-10-18 16:47:23 INFO BinaryLogClient - Connected to <**aurora instance**> at mysql-bin-changelog.009885/97551300 (sid:1729179703, cid:8700624) 18/10/2024 22:17:23.555 2024-10-18 16:47:23 INFO BinlogConnectorReplicator - Binlog connected.

There was no restart . And as you can see from the logs , binlog also reconnected . But the binlogs were not flowing into maxwell and hence causing no data in kafka topic .

We are using maxwell version : 1.40.0 .

This was resolved after redeploying maxwell :

2024-10-19 08:10:53 INFO Maxwell - Maxwell v1.40.0 is booting (MaxwellCustomKafkaProducer), starting at Position[BinlogPosition[mysql-bin-changelog.009885:97907248], lastHeartbeat=0]

This issue ( binlogs not flowing ) was not detected soon and caused incident due to dataloss ( maxwell was redeployed after the incident was opened ) .