Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.
Bug 332868 - [JUnit] Run As Junit causes kernel panic in OSX Leopard JDK 1.6
Summary: [JUnit] Run As Junit causes kernel panic in OSX Leopard JDK 1.6
Status: RESOLVED NOT_ECLIPSE
Alias: None
Product: JDT
Classification: Eclipse Project
Component: UI (show other bugs)
Version: 3.7   Edit
Hardware: Macintosh Mac OS X - Carbon (unsup.)
: P3 critical (vote)
Target Milestone: ---   Edit
Assignee: JDT-UI-Inbox CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-12-17 12:10 EST by Peter Lynch CLA
Modified: 2011-01-06 05:51 EST (History)
3 users (show)

See Also:

Attachments
First panic caused by running a Junit test (5.34 KB, text/plain)
2010-12-17 12:14 EST, Peter Lynch CLA 		no flags Details
Second panic caused by running a JUnit test (5.59 KB, text/plain)
2010-12-17 12:15 EST, Peter Lynch CLA 		no flags Details
Eclipse configuration (672.95 KB, text/plain)
2010-12-17 12:22 EST, Peter Lynch CLA 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Peter Lynch CLA 2010-12-17 12:10:13 EST 
Build Identifier: 

Twice now Eclipse has triggered a kernel panic on my OSX Leopard machine when I have tried to run a JUnit test.

I will attach excerpts from my panic reporter logs.



Reproducible: Sometimes

Steps to Reproduce:
1. Open a JUnit 4 test case
2. Right Click in Editor window and choose Run As --> JUnit Test
3. Sometimes a kernel panic is triggered requiring a restart of the machine.
Comment 1 Peter Lynch CLA 2010-12-17 12:14:48 EST 
Created attachment 185444 [details]
First panic caused by running a Junit test
Comment 2 Peter Lynch CLA 2010-12-17 12:15:25 EST 
Created attachment 185445 [details]
Second panic caused by running a JUnit test
Comment 3 Olivier Thomann CLA 2010-12-17 12:17:03 EST 
Moving to JDT/UI.
A build id and a jvm crash log would be helpful.
Comment 4 Peter Lynch CLA 2010-12-17 12:22:11 EST 
Created attachment 185446 [details]
Eclipse configuration
Comment 5 Peter Lynch CLA 2010-12-17 12:22:58 EST 
Eclipse Java EE IDE for Web Developers.

Version: Helios Service Release 1
Build id: 20100917-0705
Comment 6 Peter Lynch CLA 2010-12-17 12:56:14 EST 
Couldn't find a jvm crash log, perhaps since this was kernel panic?

(In reply to comment #3)
> Moving to JDT/UI.
> A build id and a jvm crash log would be helpful.
Comment 7 Dani Megert CLA 2011-01-03 06:39:23 EST 
>3. Sometimes a kernel panic is triggered requiring a restart of the machine.
How often does it happen? Does the same test sometimes work and sometimes crash?
Comment 8 Peter Lynch CLA 2011-01-03 11:11:40 EST 
(In reply to comment #7)
> >3. Sometimes a kernel panic is triggered requiring a restart of the machine.
> How often does it happen? Does the same test sometimes work and sometimes
> crash?

It happened twice. I am not sure if it was the same test class or method being executed - regardless the unit test itself would have been very simple and not doing much. I think one scenario was I already ran the a Junit test class ( Junit 4.x based) and test methods had failed. After making code changes and rerunning the tests, then I got kernel panic. ( OSX grey screen of death).

As you might expect after it happening twice, I didn't want to risk it happening again so have not tried lately. I had to switch to run tests from cmd line outside eclipse as a safe workaround.
Comment 9 Markus Keller CLA 2011-01-06 05:51:24 EST 
Looks like a bug in LittleSnitch:

Backtrace terminated-invalid frame pointer 0x3ffce0
      Kernel loadable modules in backtrace (with dependencies):
         at.obdev.nke.LittleSnitch(2.0.38)@0x82ffc000->0x83026fff

Please update LittleSnitch to the latest version. If that doesn't help, try uninstalling it.

Implementation note:
The JUnit launching code opens the same port twice in quick succession (bug 322266 comment 1), which could trigger a bug in LittleSnitch.