[
https://issues.apache.org/jira/browse/SOLR-17368?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17870287#comment-17870287
]
Matthew Biscocho commented on SOLR-17368:
-----------------------------------------
Ah apologies again for breaking tests... I think I figured it out. The issue is
because of my dummy metric `dummyMetrics.2xx-responses` creating a duplicate
metric of `org.eclipse.jetty.server.handler.DefaultHandler.2xx-responses` that
is leaking between tests and it creating a duplicate
`solr_metrics_jetty_response_total\{status="2xx"}` prometheus metric. Oversight
and mistake on my part. I will create a PR to resolve this. Is there a way I
can run the CI multiple times before committing? I want to be sure this test
isn't brittle to avoid this happening a 3rd time.
> TestPrometheusResponseWriter redesign
> -------------------------------------
>
> Key: SOLR-17368
> URL: https://issues.apache.org/jira/browse/SOLR-17368
> Project: Solr
> Issue Type: Test
> Security Level: Public(Default Security Level. Issues are Public)
> Reporter: Chris M. Hostetter
> Priority: Major
> Labels: pull-request-available
> Fix For: 9.7
>
> Time Spent: 50m
> Remaining Estimate: 0h
>
> TestPrometheusResponseWriter currently fails in 100% of all jenkins builds
> due to how it is designed, and depending on what other tests may run before
> it in the same JVM.
> This problem only affects the test, not the functionality of the underlying
> code.
> See SOLR-10654 for background discussions of the problems with this test, and
> options for improving it's design relative to it's purpose.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
