TestIndexWriterOnJRECrash suffers from JDK-8047340 - needs workarround [LUCENE-6036]

[asfimport]

Similar to issues uncovered in SOLR-6387, TestIndexWriterOnJRECrash does some forking which can hit JDK-8047340 on the turkish locale...

   [junit4] Suite: org.apache.lucene.index.TestIndexWriterOnJRECrash
   [junit4]   2> NOTE: download the large Jenkins line-docs file by running 'ant get-jenkins-line-docs' in the lucene directory.
   [junit4]   2> NOTE: reproduce with: ant test  -Dtestcase=TestIndexWriterOnJRECrash -Dtests.method=testNRTThreads -Dtests.seed=B2D360EA192CA242 -Dtests.multiplier=2 -Dtests.nightly=true -Dtests.slow=true -Dtests.linedocsfile=/home/jenkins/lucene-data/enwiki.random.lines.txt -Dtests.locale=tr -Dtests.timezone=Africa/Monrovia -Dtests.asserts=true -Dtests.file.encoding=ISO-8859-1
   [junit4] ERROR   0.04s J0 | TestIndexWriterOnJRECrash.testNRTThreads <<<
   [junit4]    > Throwable #1: java.lang.Error: posix_spawn is not a supported process launch mechanism on this platform.
   [junit4]    >    at __randomizedtesting.SeedInfo.seed([B2D360EA192CA242:290A74F158D7B429]:0)
   [junit4]    >    at java.lang.UNIXProcess$1.run(UNIXProcess.java:111)
   [junit4]    >    at java.lang.UNIXProcess$1.run(UNIXProcess.java:93)
   [junit4]    >    at java.security.AccessController.doPrivileged(Native Method)
   [junit4]    >    at java.lang.UNIXProcess.<clinit>(UNIXProcess.java:91)
   [junit4]    >    at java.lang.ProcessImpl.start(ProcessImpl.java:130)
   [junit4]    >    at java.lang.ProcessBuilder.start(ProcessBuilder.java:1028)
   [junit4]    >    at org.apache.lucene.index.TestIndexWriterOnJRECrash.forkTest(TestIndexWriterOnJRECrash.java:113)
   [junit4]    >    at org.apache.lucene.index.TestIndexWriterOnJRECrash.testNRTThreads(TestIndexWriterOnJRECrash.java:59)
   [junit4]    >    at java.lang.Thread.run(Thread.java:745)

https://bugs.openjdk.java.net/browse/JDK-8047340

---

Migrated from LUCENE-6036 by Chris M. Hostetter (@hossman), resolved Oct 30 2014 Attachments: LUCENE-6036.patch Linked issues:

• SOLR-6387

[asfimport]

Uwe Schindler (@uschindler) (migrated from JIRA)

Hi Hoss, can you add the same assumeTrue/False ike we did in Solr. Once this is fixed in the JDK we can remove it, but for now just let us ignore the test in Turkey :-)

[asfimport]

Uwe Schindler (@uschindler) (migrated from JIRA)

Here the patch.

[asfimport]

Uwe Schindler (@uschindler) (migrated from JIRA)

https://twitter.com/UweSays/status/501425093613207552

[asfimport]

ASF subversion and git services (migrated from JIRA)

Commit 1635549 from @uschindler in branch 'dev/trunk' https://svn.apache.org/r1635549

LUCENE-6036: Prevent this test failing on turkish default locale

[asfimport]

ASF subversion and git services (migrated from JIRA)

Commit 1635550 from @uschindler in branch 'dev/branches/branch_5x' https://svn.apache.org/r1635550

Merged revision(s) 1635549 from lucene/dev/trunk: LUCENE-6036: Prevent this test failing on turkish default locale

[asfimport]

Anshum Gupta (@anshumg) (migrated from JIRA)

Bulk close after 5.0 release.

[asfimport]

Dawid Weiss (@dweiss) (migrated from JIRA)

I think we should reopen this issue as it still affects the codebase (there are assumptions in the code that explicitly exclude when Turkish locale is set.

[asfimport]

Dawid Weiss (@dweiss) (migrated from JIRA)

I also can't change the 'resolution' and it's definitely not "fixed".

[asfimport]

Chris M. Hostetter (@hossman) (migrated from JIRA)

i think that's a quirk of the workflow? after re-opening? not sure if there is a way to remove the "resolution" once it's set, even though it's been re-opened.

[asfimport]

Dawid Weiss (@dweiss) (migrated from JIRA)

Strange... I don't even see the workflow button on this issue. I think I lost touch with Jira's changes. ;)