lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Michael McCandless <luc...@mikemccandless.com>
Subject Re: [JENKINS] Lucene-trunk-Linux-Java7-64-test-only - Build # 10035 - Failure!
Date Wed, 17 Oct 2012 17:59:10 GMT
On Wed, Oct 17, 2012 at 1:26 PM, Dawid Weiss
<dawid.weiss@cs.put.poznan.pl> wrote:
>> java7). I even tried tacking on -Dtests.jvms=1 -Dtests.dups=20 in case
>> its a testrunner leak (this one makes lots of threads).
>
> Shouldn't be the runner's problem anymore (I fixed the previous
> issue). Although proving me wrong by showing an example (heapdump)
> will make me change my opinion of course ;)

Curiously ... the other OOME (repros easily for me: "ant test
-Dtestcase=TestIndexWriter -Dtests.method=testRotatingFieldNames
-Dtests.seed=38ACAB42219DC17B -Dtests.slow=true -Dtests.locale=ar_LY
-Dtests.timezone=Atlantic/Faroe -Dtests.file.encoding=UTF-8") repros
if I use randomized runner but not if I use JUnit's runner.

But in the heap dump nothing stands out from randomized runner ... I
see lots of stuff in BlockTermsWriter that I'll fix ...

Still, it's spooky, because I don't think anything has changed in this
test or in BlockTermsWriter ... so they didn't provoke OOMEs before
but now they do ... so I think something else is holding onto RAM.  Or
maybe we just hadn't hit the magical seed in the past ... just seems
like suddenly we're getting many OOMEs.

Mike McCandless

http://blog.mikemccandless.com

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


Mime
View raw message