lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Michael McCandless <luc...@mikemccandless.com>
Subject Re: Why would test timeout not kick in...?
Date Fri, 30 Jan 2015 14:37:05 GMT
Thanks Uwe & Dawid, next time I'll try try to SIGHUP it.

I never knew about this -DSelfDestructTime=30!

Mike McCandless

http://blog.mikemccandless.com


On Mon, Jan 26, 2015 at 3:21 PM, Dawid Weiss
<dawid.weiss@cs.put.poznan.pl> wrote:
> If you encounter a situation like this, please stack dump the hung
> JVM, ok? (send a SIGHUP signal to it). If you can't provoke the JVM to
> dump its stack then it's very likely a JVM error. Otherwise I'll have
> something to debug.
>
> An alternative solution to force-kill a forked JVM (Oracle only) is to
> pass the magic switch to it:
>
> -DSelfDestructTimer=30
>
> The number is in minutes; from JDK sources:
>
> product(intx, SelfDestructTimer, 0,                                       \
>           "Will cause VM to terminate after a given time (in minutes) "     \
>           "(0 means off)")
>
> Dawid
>
> On Mon, Jan 26, 2015 at 8:42 PM, Uwe Schindler <uwe@thetaphi.de> wrote:
>> Hi,
>>
>> This happens in most cases under OOM situations. In that case the test runner loses
control and is unable to shut down. In this case it could be something different, because
you still see a "test method" in the hearbeat. On OOM situations in most cases you see just
the test case name and no method. We have that quite often with Solr tests on Policeman Jenkins,
too. If you want to be sure that a build is aborted, you can also set a maximum timeout for
the whole build in Jenkins. Jenkins will then "kill -9" the whole process structure it launched.
Please note that Jenkins measures the whole build time, so give enough buffer.
>>
>> Uwe
>>
>> -----
>> Uwe Schindler
>> H.-H.-Meier-Allee 63, D-28213 Bremen
>> http://www.thetaphi.de
>> eMail: uwe@thetaphi.de
>>
>>
>>> -----Original Message-----
>>> From: Michael McCandless [mailto:lucene@mikemccandless.com]
>>> Sent: Monday, January 26, 2015 7:10 PM
>>> To: Lucene/Solr dev
>>> Subject: Why would test timeout not kick in...?
>>>
>>> This test (TestCompressingTermVectorsFormat.testClone) just kept
>>> HEARTBEAT-ing for 2 days:
>>>
>>>     http://build-eu-
>>> 00.elasticsearch.org/job/lucene_linux_java8_64_test_only/26953/console
>>>
>>> The test class / super classes are not annotated with longer timeouts ...
>>>
>>> Shouldn't it have timed out at 7200 seconds?
>>>
>>> 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
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: dev-help@lucene.apache.org
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: dev-help@lucene.apache.org
>

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


Mime
View raw message