[
https://issues.apache.org/jira/browse/HBASE-15425?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15189547#comment-15189547
]
Ashish Singhi commented on HBASE-15425:
---------------------------------------
Actually that abort is a graceful shutdown. In our setup what we have seen is,
client is able to get the response and the RS is getting shut down. But yes it
is possible that before the client get response RS can get aborted.
bq. Should you avoid the RS from getting aborted?
Yes, I was thinking on that as well, that RS abort is not really required when
WAL append is failing. But wanted to check all the scenarios around it before
raising a jira. I will raise one tomorrow(IST) and address that.
> Failing to write bulk load event marker in the WAL is ignored
> -------------------------------------------------------------
>
> Key: HBASE-15425
> URL: https://issues.apache.org/jira/browse/HBASE-15425
> Project: HBase
> Issue Type: Bug
> Affects Versions: 1.3.0
> Reporter: Ashish Singhi
> Assignee: Ashish Singhi
> Fix For: 2.0.0, 1.3.0, 1.4.0
>
> Attachments: HBASE-15425.patch, HBASE-15425.v1.patch
>
>
> During LoadIncrementalHFiles process if we fail to write the bulk load event
> marker in the WAL, it is ignored. So this will lead to data mismatch issue in
> source and peer cluster in case of bulk loaded data replication scenario.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
