[
https://issues.apache.org/jira/browse/HBASE-14949?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15052248#comment-15052248
]
Heng Chen commented on HBASE-14949:
-----------------------------------
{quote}
And I checked the code again, for increment and append, we first get the row
from region, do increment or append, and log the entire cell out as WAL, so it
is also safe to replay it multiple times.
{quote}
Thanks [~stack] [~Apache9] for your reply. So there is no need for this issue.
{quote}
We should test if things go right when replaying same WAL entry multiple times.
And I think we could do it along with the fixing of WAL file name conflict
after splitting.
{quote}
OK, it seems we could do some changes base on the first patch.
> Skip duplicate entries when replay WAL.
> ---------------------------------------
>
> Key: HBASE-14949
> URL: https://issues.apache.org/jira/browse/HBASE-14949
> Project: HBase
> Issue Type: Sub-task
> Reporter: Heng Chen
> Attachments: HBASE-14949.patch, HBASE-14949_v1.patch
>
>
> As HBASE-14004 design, there will be duplicate entries in different WAL. It
> happens when one hflush failed, we will close old WAL with 'acked hflushed'
> length, then open a new WAL and write the unacked hlushed entries into it.
> So there maybe some overlap between old WAL and new WAL.
> We should skip the duplicate entries when replay. I think it has no harm to
> current logic, maybe we do it first.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
