[
https://issues.apache.org/jira/browse/HBASE-10557?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13903788#comment-13903788
]
Feng Honghua commented on HBASE-10557:
--------------------------------------
Update:
# DroppedSnapshotException(as Throwable) thrown during hlog-replay is
eventually handled in OpenRegionHandler by failing the region opening of which
the hlog-replay is a sub-process, the region is not successfully opened in the
regionserver, and master will re-assign this region, no problem regarding
data-loss.
# DroppedSnapshotException(as Throwable) thrown during non-abort region-close
is eventually handled in CloseRegionHandler by aborting the hosting
regionserver, also no risk of data-loss.
According to above, this jira is a false alarm and can be closed as something
like 'won't fix' or 'not a problem'. Sorry :-)
> DroppedSnapshotException is not handled properly for flush triggered by
> hlog-replay and non-abort region close
> --------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-10557
> URL: https://issues.apache.org/jira/browse/HBASE-10557
> Project: HBase
> Issue Type: Bug
> Components: regionserver
> Reporter: Feng Honghua
> Assignee: Feng Honghua
> Priority: Critical
>
> Flush triggered by hlog-replay(replayRecoveredEdits) and
> region-close(non-abort close) get processed directly by region without
> putting flush entry into flushQueue, hence not handled by MemStoreFlusher, So
> DroppedSnapshotException emitted from internalFlushcache is not handled
> properly
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
