[
https://issues.apache.org/jira/browse/HBASE-8028?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13620665#comment-13620665
]
Himanshu Vashishtha commented on HBASE-8028:
--------------------------------------------
[~jxiang] Very good points.
As Ted said, readpoint is not changed. The readpoint which is being used for
the rollback is not committed yet, so none of the scanner is reading it at this
point.
For the snapshots, the patch needs to be altered a bit. Basically, if the kv is
present in snapshot then it will not be in memstore kv, and vice versa.
So, while deleting the new kv, if it is deleted from snapshot, then the old kv
should be only added to the snapshot. Similarly, if the new kv is deleted from
memstore's kv, then the old kv should be added to the memstore's kv.
I'll upload a revised patch.
> Append, Increment: Adding rollback support
> ------------------------------------------
>
> Key: HBASE-8028
> URL: https://issues.apache.org/jira/browse/HBASE-8028
> Project: HBase
> Issue Type: Bug
> Components: regionserver
> Affects Versions: 0.94.5
> Reporter: Himanshu Vashishtha
> Assignee: Himanshu Vashishtha
> Fix For: 0.95.0
>
> Attachments: HBase-8028-v1.patch, HBase-8028-v2.patch,
> HBase-8028-with-Increments-v1.patch, HBase-8028-with-Increments-v2.patch
>
>
> In case there is an exception while doing the log-sync, the memstore is not
> rollbacked, while the mvcc is _always_ forwarded to the writeentry created at
> the beginning of the operation. This may lead to scanners seeing results
> which are not synched to the fs.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
