[
https://issues.apache.org/jira/browse/HBASE-17482?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15828239#comment-15828239
]
Allan Yang commented on HBASE-17482:
------------------------------------
Just looked code in branch-1, it doesn't have this issue. It looks like when
HBASE-16698 was ported to master branch, some code is lost.
This is the critical line in branch-1
{code}
boolean updateSeqId = isInReplay
|| batchOp.getMutation(i).getDurability() == Durability.SKIP_WAL ||
mvccPreAssign;
{code}
> mvcc mechanism failed when using mvccPreAssign
> ----------------------------------------------
>
> Key: HBASE-17482
> URL: https://issues.apache.org/jira/browse/HBASE-17482
> Project: HBase
> Issue Type: Bug
> Affects Versions: 2.0..
> Reporter: Allan Yang
> Assignee: Allan Yang
> Priority: Critical
> Attachments: HBASE-17482.patch, HBASE-17482.v2.patch,
> HBASE-17482.v3.patch
>
>
> If mvccPreAssign and ASYNC_WAL is used, then cells may have been commited to
> memstore before append thread can stamp seqid to them. The unit test shows
> everything.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
