[
https://issues.apache.org/jira/browse/FLINK-16686?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17558119#comment-17558119
]
Alexis Sarda-Espinosa commented on FLINK-16686:
-----------------------------------------------
It seems I spoke to soon, something definitely changed but I still get an
exception, it's just that it's thrown after approximately 1 hour instead of 30
seconds, not sure why.
> [State TTL] Make user class loader available in native RocksDB compaction
> thread
> --------------------------------------------------------------------------------
>
> Key: FLINK-16686
> URL: https://issues.apache.org/jira/browse/FLINK-16686
> Project: Flink
> Issue Type: Bug
> Components: Runtime / State Backends
> Affects Versions: 1.8.0, 1.11.3, 1.13.0, 1.12.3
> Reporter: Andrey Zagrebin
> Priority: Minor
> Labels: auto-deprioritized-major, auto-deprioritized-minor
>
> The issue is initially reported
> [here|https://stackoverflow.com/questions/60745711/flink-kryo-serializer-because-chill-serializer-couldnt-be-found].
> The problem is that the java code of Flink compaction filter is called from
> RocksDB native C++ code. It is called in the context of the native compaction
> thread. RocksDB has utilities to create java Thread context for the Flink
> java callback. Presumably, the Java thread context class loader is not set at
> all and if it is queried then it produces NullPointerException.
> The provided report enabled a list state with TTL. The compaction filter has
> to deserialise elements to check expiration. The deserialiser relies on Kryo
> which queries the thread context class loader which is expected to be the
> user class loader of the task but turns out to be null.
> We should investigate how to pass the user class loader to the compaction
> thread of the list state with TTL.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
