[
https://issues.apache.org/jira/browse/HBASE-10196?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13851731#comment-13851731
]
Jean-Marc Spaggiari commented on HBASE-10196:
---------------------------------------------
Thanks a lot [~zjushch] for fixing this!
Few comments.
{code}
+ if (hbi.isMerged()) {
+ // This region has already been merged, the remaining hdfs file will be
+ // cleaned by CatalogJanitor later
+ hbi.setSkipChecks(true);
+ return;
+ }
{code}
Should we not set a timeout? Like, if the region isMerged but is there for more
than xx minutes/hours, then maybe the CatalogJanitor is failing to remove it so
we should pop an alert?
{code}
+ if (mergeRegion != null) {
{code}
Can this really occurs? I'm not 100% sure. If it's null, does it mean there is
something wrong and we should log an alert/warning/debug/trace?
Overall, big +1 from me!
> Enhance HBCK to understand the case after online region merge
> -------------------------------------------------------------
>
> Key: HBASE-10196
> URL: https://issues.apache.org/jira/browse/HBASE-10196
> Project: HBase
> Issue Type: Bug
> Reporter: chunhui shen
> Assignee: chunhui shen
> Fix For: 0.96.2, 0.98.1, 0.99.0
>
> Attachments: hbase-10196-trunk.patch
>
>
> HBCK will report an error now after the online merge,
> because the files of merging regions still remain on HDFS which will be
> cleaned by CatalogJanitor later.
> ERROR: Region { meta => null, hdfs =>
> hdfs://hbasetest1:9000/hbase/data/default/dns/c6569a72cc3c2750d14976ab85f02315,
> deployed => } on HDFS, but not listed in hbase:meta or deployed on any
> region server
--
This message was sent by Atlassian JIRA
(v6.1.4#6159)
