[
https://issues.apache.org/jira/browse/HBASE-6784?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13464828#comment-13464828
]
Lars Hofhansl commented on HBASE-6784:
--------------------------------------
@Ram:
Most Unix's have good timers (I know Linux has), so it is rare that the two
Puts here receive the same timestamp (but still possible, so the test was still
bad).
The last time I played with timers on Windows I found that the timer resolution
there is about 10ms (but that was many years ago and might be better now).
> TestCoprocessorScanPolicy is sometimes flaky when run locally
> -------------------------------------------------------------
>
> Key: HBASE-6784
> URL: https://issues.apache.org/jira/browse/HBASE-6784
> Project: HBase
> Issue Type: Bug
> Reporter: ramkrishna.s.vasudevan
> Assignee: Lars Hofhansl
> Priority: Minor
> Fix For: 0.94.2, 0.96.0
>
> Attachments: 6784.txt
>
>
> The problem is not seen in jenkins build.
> When we run TestCoprocessorScanPolicy.testBaseCases locally or in our
> internal jenkins we tend to get random failures. The reason is the 2 puts
> that we do here is sometimes getting the same timestamp. This is leading to
> improper scan results as the version check tends to skip one of the row
> seeing the timestamp to be same. Marking this as minor. As we are trying to
> solve testcase related failures just raising this incase we need to resolve
> this also.
> For eg,
> Both the puts are getting the time
> {code}
> time 1347635287360
> time 1347635287360
> {code}
--
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
