[
https://issues.apache.org/jira/browse/KAFKA-9062?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16954994#comment-16954994
]
Sophie Blee-Goldman commented on KAFKA-9062:
--------------------------------------------
I think this is challenging to solve in a holistic way without a large scale
architectural change. Personally I'd vote to focus on a short term patch for
the specific case users seem to be hitting while keeping this is mind when
discussing longer term revamping of Streams.
That's an interesting idea, one other thought I had was just to allow the bulk
loading strategy to be more configurable as it seems to work well for some
users and clearly not so well for others. This could actually be related to
KAFKA-8627 as I believe pre-sorting may greatly improve the compaction times
> Handle stalled writes to RocksDB
> --------------------------------
>
> Key: KAFKA-9062
> URL: https://issues.apache.org/jira/browse/KAFKA-9062
> Project: Kafka
> Issue Type: Bug
> Components: streams
> Reporter: Sophie Blee-Goldman
> Priority: Major
>
> RocksDB may stall writes at times when background compactions or flushes are
> having trouble keeping up. This means we can effectively end up blocking
> indefinitely during a StateStore#put call within Streams, and may get kicked
> from the group if the throttling does not ease up within the max poll
> interval.
> Example: when restoring large amounts of state from scratch, we use the
> strategy recommended by RocksDB of turning off automatic compactions and
> dumping everything into L0. We do batch somewhat, but do not sort these small
> batches before loading into the db, so we end up with a large number of
> unsorted L0 files.
> When restoration is complete and we toggle the db back to normal (not bulk
> loading) settings, a background compaction is triggered to merge all these
> into the next level. This background compaction can take a long time to merge
> unsorted keys, especially when the amount of data is quite large.
> Any new writes while the number of L0 files exceeds the max will be stalled
> until the compaction can finish, and processing after restoring from scratch
> can block beyond the polling interval
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
