[
https://issues.apache.org/jira/browse/KYLIN-5029?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17392383#comment-17392383
]
ASF subversion and git services commented on KYLIN-5029:
--------------------------------------------------------
Commit 133bfb8020ebc50fe712f0745e961c5dea51f072 in kylin's branch
refs/heads/main from Congling Xia
[ https://gitbox.apache.org/repos/asf?p=kylin.git;h=133bfb8 ]
KYLIN-5029 set metrics refresh interval to a smaller value to make sure tests
always passed (#1683)
* KYLIN-5029 set metrics refresh interval to a smaller value to make sure tests
always passed
* refine
> metrics test may fail because the cache is not refreshed
> --------------------------------------------------------
>
> Key: KYLIN-5029
> URL: https://issues.apache.org/jira/browse/KYLIN-5029
> Project: Kylin
> Issue Type: Improvement
> Components: Tools, Build and Test
> Reporter: Congling Xia
> Assignee: Congling Xia
> Priority: Major
>
> Tests in _org.apache.kylin.rest.metrics.QueryMetricsTest_ may fail in current
> CI processes. There are some "Thread.sleep(..)" in the test code to wait for
> metrics cache being refreshed.
> After digging into the Hadoop metrics implementation, I find that the default
> refreshing interval is 10 seconds. So test thread should sleep for more than
> 10 seconds to may sure metrics are up-to-date. I checked the sleep intervals
> in the test code and they are not greater than 10000 (ms). That's why CI
> processes fail accidentally.
> It is better to configure metrics refreshing interval to a smaller value.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
