[
https://issues.apache.org/jira/browse/HBASE-22075?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16869837#comment-16869837
]
Sean Busbey commented on HBASE-22075:
-------------------------------------
I had a reproduction attempt timeout because something went wonky that bubbled
up to compactions failing. in v1 when that happens we immediately repeat the
request instead of waiting. that eventually put enough pressure on things that
my attempts to write failed, which just exited the data writing thread and left
the test to wait for the full timeout.
test-only v2
- move to java.util.concurrent Executor to handle periodic tasks and waiting
for an async task to be done
- if the data writing thread gets an error end the test as a failure right away
> Potential data loss when MOB compaction fails
> ---------------------------------------------
>
> Key: HBASE-22075
> URL: https://issues.apache.org/jira/browse/HBASE-22075
> Project: HBase
> Issue Type: Bug
> Components: mob
> Affects Versions: 2.1.0, 2.0.0, 2.0.1, 2.1.1, 2.0.2, 2.0.3, 2.1.2, 2.0.4,
> 2.1.3
> Reporter: Vladimir Rodionov
> Assignee: Vladimir Rodionov
> Priority: Critical
> Labels: compaction, mob
> Fix For: 2.0.6, 2.2.1, 2.1.6
>
> Attachments: HBASE-22075-v1.patch, HBASE-22075-v2.patch,
> HBASE-22075.test-only.0.patch, HBASE-22075.test-only.1.patch,
> HBASE-22075.test-only.2.patch, ReproMOBDataLoss.java
>
>
> When MOB compaction fails during last step (bulk load of a newly created
> reference file) there is a high chance of a data loss due to partially loaded
> reference file, cells of which refer to (now) non-existent MOB file. The
> newly created MOB file is deleted automatically in case of a MOB compaction
> failure, but some cells with the references to this file might be loaded to
> HBase.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
