Bug 440825 - Eclipse Juno 32-bit keeps crashing without warning
Summary: Eclipse Juno 32-bit keeps crashing without warning
Status: CLOSED WONTFIX
Alias: None
Product: Platform
Classification: Eclipse Project
Component: IDE (show other bugs)
Version: 4.2.2   Edit
Hardware: PC Windows 7
: P3 critical (vote)
Target Milestone: ---   Edit
Assignee: Platform-UI-Inbox CLA
QA Contact:
URL:
Whiteboard:
Keywords: needinfo
Depends on:
Blocks:
 
Reported: 2014-07-30 19:50 EDT by A B CLA
Modified: 2014-08-06 18:45 EDT (History)
1 user (show)

See Also:


Attachments
One of the Eclipse Crashes Pop-ups and Stacktrace (151.41 KB, application/octet-stream)
2014-07-30 19:50 EDT, A B CLA
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description A B CLA 2014-07-30 19:50:17 EDT
Created attachment 245563 [details]
One of the Eclipse Crashes Pop-ups and Stacktrace

In our organisation, we are trying to move from RAD to Eclipse.  We have got it to work from technical perspective and the process has been set-up to install Eclipse Juno 32-bit JEE version, WAS for Developers version, CC and CQ. The tests have been working quite well.

But Developers have been complaining that their Eclipse crashes in the middle of their work without any warning.  It makes them lose their work and also decreases their productivity.   

We are using the location of 32-bit websphere for Developers versions 1.7 JDK location as the -vm paramter in the eclipse.ini.

One of the developers sent me the attached error from one such crashes.

Please let me know, if I can provide any mor einformation that may help.
Comment 1 Daniel Rolka CLA 2014-07-31 03:27:13 EDT
The issue can be related to the WAS Dev plugin, I see it in the following stack trace:

com.ibm.etools.rpe.internal.ui.DesignPaneController$12$1.run(DesignPaneController.java:1270)at org.eclipse.core.internal.jobs.Worker.run(Worker.java:53)

Have you tried to use Kepler or Luna and you got the same issue? Personally I would migrate to the Kepler or Luna that are more stable than Juno in my opinion

The proper packages you can find here - http://eclipse.org/downloads/

Daniel
Comment 2 Paul Webster CLA 2014-07-31 15:05:18 EDT
If it crashed, it should have produced a VM stack trace (hs_err_<pid>.log in the case of the Oracle JVM, not sure of the IBM version).  Please attach that here.

PW
Comment 3 A B CLA 2014-08-02 08:38:19 EDT
(In reply to Daniel Rolka from comment #1)

> Have you tried to use Kepler or Luna and you got the same issue? Personally I 
> would migrate to the Kepler or Luna that are more stable than Juno in my opinion

> The proper packages you can find here - http://eclipse.org/downloads/

> Daniel

My problem with trying Kepler or Luna is that we use ClearCase and the IBM provided clearcase plug-in works with Juno.  Their support document specifically mentions that it will fail with Luna.  I did not find any specifcs about making it work with Kepler.  But I can try that.

Are you aware of Clearcase plug-in updates for Luna that I could try?
Comment 4 A B CLA 2014-08-02 08:40:28 EDT
(In reply to Paul Webster from comment #2)
> If it crashed, it should have produced a VM stack trace (hs_err_<pid>.log in
> the case of the Oracle JVM, not sure of the IBM version).  Please attach
> that here.

PW

I did not find these logs with IBM JVM when the crashes happened.  Is there something else I should look for?
Comment 5 A B CLA 2014-08-06 18:45:54 EDT
(In reply to A B from comment #4)

We finally figure out the cause of the crashes.  The JSP and xhtml pages had "Rich Page Editor" as the default editor.  What we observed that when you open either a jsp or an xhtml page with Rich Page Editor, the memory just keeps climbing, even if you are not doing anything in the project and ultimately Eclipse just crashes.  We changed the default editor to Web Page Editor and the eclipse on different machines haven't cashed for last two days ( normally, it would happen 4-5 times / day on each of those machines )