[jira] [Commented] (SOLR-9111) HdfsCollectionsAPIDistributedZkTest stalled for over 75 hours

["Dawid Weiss (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SOLR-9111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15283936#comment-15283936 ] 

Dawid Weiss commented on SOLR-9111:
-----------------------------------

I think all tests do have a timeout (if they inherit from LuceneTestCase). It could be overriden from command-line as:
{code}
-Dtests.suiteTimeout=5000!
{code}

this means 5000 millis, forced over all suites.

Note that hung JVMs that *do not* timeout are typically JVMs that are not responsive or frozen for some other reason. This has happened in the past (the timeouting is implemented on the forked JVM side, to allow IDEs, etc. to benefit from it as they don't fork a separate JVM.

> HdfsCollectionsAPIDistributedZkTest stalled for over 75 hours
> -------------------------------------------------------------
>
>                 Key: SOLR-9111
>                 URL: https://issues.apache.org/jira/browse/SOLR-9111
>             Project: Solr
>          Issue Type: Bug
>          Components: Tests
>            Reporter: Steve Rowe
>            Priority: Minor
>
> In [https://builds.apache.org/job/Lucene-Solr-NightlyTests-6.x/63], HdfsCollectionsAPIDistributedZKTest stalled for 272347s - over 75 hours - before I killed the job.
> I think it should have a timeout, or maybe nightly tests should have a global timeout?
> Except for the monster tests, it seems to me that 24 hours is a reasonable limit on the full suite.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org