[
https://issues.apache.org/jira/browse/HDFS-11630?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15959997#comment-15959997
]
Hanisha Koneru commented on HDFS-11630:
---------------------------------------
[~arpitagarwal], thanks for the review.
bq. Also instead of a single call, you can probably use a loop like you have in
the other two test cases.
The FutureCallBack used in this test is mock FutureCallBack object. So, we
would not have the exit condition for loop. The other option is to not use a
mock FutureCallBack.
bq. Another suggestion is to make the timeouts less aggressive. We can set it
them to e.g. 120 seconds instead of 1 or 2 seconds to reduce spurious timeouts
when testing on overloaded VMs.
Would 120 seconds not be too long to test for timeouts. Since we make the dummy
checkable wait for the timeout, could we not have shorter timeouts (like 2
seconds) and spurious timeouts would not have negative impact on the tests.
What do you think?
> TestThrottledAsyncCheckerTimeout fails intermittently in Jenkins builds
> -----------------------------------------------------------------------
>
> Key: HDFS-11630
> URL: https://issues.apache.org/jira/browse/HDFS-11630
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: hdfs
> Reporter: Hanisha Koneru
> Assignee: Hanisha Koneru
> Attachments: HDFS-11630.001.patch
>
>
> TestThrottledAsyncCheckerTimeout#testDiskCheckTimeoutInvokesOneCallbackOnly
> fails intermittently in Jenkins builds.
> We need to wait for disk checker timeout to callback the
> FutureCallBack#onFailure.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
