[
https://issues.apache.org/jira/browse/PHOENIX-5612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alex Batyrshin updated PHOENIX-5612:
------------------------------------
Description: Upsert Select running via Parallel Iteration Factory does not
commit last chunk of data in iteration threads. Last chunk (MutationState)
returned to main thread instead. This final MutationState from all threads is
then joining to connection.getMutatioState. This could result that
connection.getMutatioState grow up and exceed limits (maxSize or maxSizeBytes).
(was: Upsert Select running via Parallel Iteration Factory does not commit
last chunk of data in iteration threads. Instead last chunk (MutationState)
returned to main thread. This final MutationState from all threads is then
joining to connection.getMutatioState. This could result that
connection.getMutatioState grow up and exceed limits (maxSize or maxSizeBytes).)
> Upsert Select exceed limits on last execution step of joining threads final
> mutationState
> -----------------------------------------------------------------------------------------
>
> Key: PHOENIX-5612
> URL: https://issues.apache.org/jira/browse/PHOENIX-5612
> Project: Phoenix
> Issue Type: Bug
> Affects Versions: 4.14.2, 4.14.3
> Reporter: Alex Batyrshin
> Priority: Major
> Attachments: PHOENIX-5612.4.x-HBase-1.4.v1.patch,
> PHOENIX-5612.4.x.000.patch, PHOENIX-5612.master.000.patch,
> commit_last_chunk.patch
>
>
> Upsert Select running via Parallel Iteration Factory does not commit last
> chunk of data in iteration threads. Last chunk (MutationState) returned to
> main thread instead. This final MutationState from all threads is then
> joining to connection.getMutatioState. This could result that
> connection.getMutatioState grow up and exceed limits (maxSize or
> maxSizeBytes).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
