[
https://issues.apache.org/jira/browse/LUCENE-4252?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13422596#comment-13422596
]
Dawid Weiss commented on LUCENE-4252:
-------------------------------------
bq. I didnt want to make all test stacktraces longer though
I honestly don't think this is such a big issue -- stack traces are for
inspecting what went wrong; if something did you look at it top-to-bottom
anyway. Some people are uncomfortable seing a long stack trace (or a nested
one); we can apply some filtering to them but I'd say let's just keep them
verbose. Better show more than needed than hide what's important.
bq. with my idea to catch static initializers.
I looked at the patch but I didn't see where this idea went; perhaps I missed
something. Anyway, not important for now, I'll revisit after I'm back home.
> Detect/Fail tests when they leak RAM in static fields
> -----------------------------------------------------
>
> Key: LUCENE-4252
> URL: https://issues.apache.org/jira/browse/LUCENE-4252
> Project: Lucene - Core
> Issue Type: Test
> Components: general/test
> Reporter: Robert Muir
> Assignee: Dawid Weiss
> Attachments: LUCENE-4252.patch, LUCENE-4252.patch
>
>
> We run our junit tests without firing up a JVM each time.
> But some tests initialize lots of stuff in @BeforeClass and don't properly
> null it out in an @AfterClass, which can cause a subsequent test in the same
> JVM to OOM, which is difficult to debug.
> Inspiration for this was me committing Mike's cool TestPostingsFormat, which
> forgot to do this: then we were seeing OOMs in several jenkins runs.
> We should try to detect these leaks in LuceneTestCase with RAMUsageEstimator
> and fail the test.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
