I can’t understand the ‘queryTimeAfterNumEntries’ parameter of the ‘cleanupInRocksdbCompactFilter’ method regarding TTL settings in Apache Flink 1.15.
Doc says:

I can’t understand the ‘queryTimeAfterNumEntries’ parameter of the ‘cleanupInRocksdbCompactFilter’ method regarding TTL settings in Apache Flink 1.15.
Doc says:

I have this hint in the logs of the my Flink 1.15 app for ‘java.time.Duration’ and ‘java.util.HashMap’:

I have a Flink 1.15 application running on AWS KDA (Managed Apache FLink). The application is being subjected to Snapshot (savepoint) on update. I got this exception (at the bottom of the message) I think because I moved a wrapped class from a ValueState (and therefore subjected to savepoint).