[
https://issues.apache.org/jira/browse/HIVE-8966?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14270282#comment-14270282
]
Alan Gates commented on HIVE-8966:
----------------------------------
The issue is that since the writer died with an unclosed batch it left the orc
file in a state where it cannot be read without the length file. So removing
the length file means any reader will fail when reading it.
The proper solution is for the compactor to stop at that partition until it has
determined all transactions in that file have committed or aborted. Then it
should compact it using the length file, but properly ignore the length file.
I'll work on the fix.
> Delta files created by hive hcatalog streaming cannot be compacted
> ------------------------------------------------------------------
>
> Key: HIVE-8966
> URL: https://issues.apache.org/jira/browse/HIVE-8966
> Project: Hive
> Issue Type: Bug
> Components: HCatalog
> Affects Versions: 0.14.0
> Environment: hive
> Reporter: Jihong Liu
> Assignee: Alan Gates
> Priority: Critical
> Fix For: 0.14.1
>
> Attachments: HIVE-8966.2.patch, HIVE-8966.3.patch, HIVE-8966.patch
>
>
> hive hcatalog streaming will also create a file like bucket_n_flush_length in
> each delta directory. Where "n" is the bucket number. But the
> compactor.CompactorMR think this file also needs to compact. However this
> file of course cannot be compacted, so compactor.CompactorMR will not
> continue to do the compaction.
> Did a test, after removed the bucket_n_flush_length file, then the "alter
> table partition compact" finished successfully. If don't delete that file,
> nothing will be compacted.
> This is probably a very severity bug. Both 0.13 and 0.14 have this issue
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
