[
https://issues.apache.org/jira/browse/HBASE-28027?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17755750#comment-17755750
]
Duo Zhang commented on HBASE-28027:
-----------------------------------
OK, sometimes it still failed with the above exception, but anyway, it could
also sometimes pass now, we have made some progress.
Let me apply the patch to all active branches and open another issue to address
the new problem.
Thanks.
> Make TestClusterScopeQuotaThrottle run faster
> ---------------------------------------------
>
> Key: HBASE-28027
> URL: https://issues.apache.org/jira/browse/HBASE-28027
> Project: HBase
> Issue Type: Sub-task
> Components: Quotas, test
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Priority: Major
>
> -The test always times out and it has several test methods.
> Let's split the test into several smaller tests, so we can find out which one
> is the criminal.-
> Finally I found that, the problem is we does not limit the operation timeout,
> just set the max retry number, but after some new improvements come in,
> sometimes we may get a 6 min sleep time and n doubt we will get a test
> timeout...
> I changed the test a bit to set operation timeout on the Table instance we
> use, so it will fail immediately when we hit the quota throttling, and now
> the tests could finish very soon.
> I think we could add another E2E tests to make sure that the refilling works
> as expected.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
