[
https://issues.apache.org/jira/browse/HBASE-9502?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13765226#comment-13765226
]
Liang Xie commented on HBASE-9502:
----------------------------------
I changed the case, to:
1) let block size to smaller, such that each block could have 2 kv at least,
else "walkForwardInSingleRow" will take effect:)
2) let the seekTo can reture -2(I just realize our internal generate faked key
algo seems more aggressive, so it's easy to repro, i'll file another jira for
that, and after that patch, you'll find without this change, the case will be
fail always)
> HStore.seekToScanner should handle magic value
> ----------------------------------------------
>
> Key: HBASE-9502
> URL: https://issues.apache.org/jira/browse/HBASE-9502
> Project: HBase
> Issue Type: Bug
> Components: regionserver, Scanners
> Affects Versions: 0.98.0, 0.95.2, 0.96.1
> Reporter: Liang Xie
> Assignee: Liang Xie
> Attachments: HBASE-9502.txt
>
>
> due to faked key, the seekTo probably reture -2, and HStore.seekToScanner
> should handle this corner case.
--
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
