eclipse 3.3.2 crashes on project right-click on java-6-openjdk and java-6-sun, but not java-1.5.0-sun (Hardy regression - libxcb.so)

Bug #217538 reported by John Greeley
52
This bug affects 1 person
Affects Status Importance Assigned to Milestone
openjdk-6 (Ubuntu)
Expired
Undecided
Unassigned

Bug Description

I can reliably reproduce this bug on Hardy with Eclipse 3.3.2, and confirm that this problem did not exist in Gutsy.

1. Set my java alternatives to java-6-openjdk or java-6-sun
2. Run my externally unpacked eclipse 3.3.2 install from /opt/eclipse
3. Wait for eclipse to finish loading
4. Right click on an open project in the workspace

Eclipse immediately terminates without any warning or explanation. This bug will appear at other times during eclipse usage that appear to be random (to me) but right-clicking on a project will crash eclipse without fail at any point during usage (always before the right-click menu even appears). There is no output in the .metadata/.log file where eclipse normally leaves crash info.

If I run eclipse from the command line I get the following error at eclipse termination:

#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0xb75eeeb5, pid=18270, tid=3074725664
#
# Java VM: OpenJDK Client VM (1.6.0-b08 mixed mode, sharing linux-x86)
# Problematic frame:
# C [libxcb.so.1+0x7eb5]
#
# An error report file with more information is saved as:
# /opt/eclipse/hs_err_pid18270.log
#
# If you would like to submit a bug report, please visit:
# https://bugs.launchpad.net/ubuntu/+source/openjdk-6/
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#
Aborted (core dumped)

I have attached the hs_err_pid18270.log file that the JVM dumped.

----

This bug does not manifest when running on the latest java-1.5.0-sun runtime.

Also, I previously experienced the "sloppy locking" problem relating to libxcb.so in other apps like Netbeans as well as Eclipse but it did not cause the application to crash like this. The "sloppy locking" issue has been fixed by recent updates, but this fatal eclipse issue remains (I cannot reproduce it with any other java application).

Revision history for this message
John Greeley (john.greeley) wrote :
Revision history for this message
John Greeley (john.greeley) wrote :

I can tentatively state that the OpenJDK updates today (6b09-0ubuntu1 - Tues April 15) have fixed this problem, we'll see whether this bug returns after a day of work usage in eclipse.

Revision history for this message
John Greeley (john.greeley) wrote :

Nope, this is still an issue. An serious one at that.

#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0xb7597eac, pid=12470, tid=3074369312
#
# Java VM: OpenJDK Client VM (1.6.0-b09 mixed mode, sharing linux-x86)
# Problematic frame:
# C [libxcb.so.1+0x7eac]
#
# If you would like to submit a bug report, please visit:
# https://bugs.launchpad.net/ubuntu/+source/openjdk-6/
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

Revision history for this message
Matthias Klose (doko) wrote :

do you see the same with the eclipse package as found in hardy?

Changed in openjdk-6:
status: New → Incomplete
Revision history for this message
John Greeley (john.greeley) wrote :

No, the same bug doesn't appear in the eclipse 3.2.2 in hardy (with limited testing)

Revision history for this message
John Greeley (john.greeley) wrote :

The newest openjdk build (6b09-0ubuntu2) changed the "problematic frame" entry, but the crash behavior hasn't changed.

#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x83ef4f7c, pid=10352, tid=3074291488
#
# Java VM: OpenJDK Client VM (1.6.0-b09 mixed mode, sharing linux-x86)
# Problematic frame:
# C 0x83ef4f7c
#
# An error report file with more information is saved as:
# /home/jgreeley/archive/hs_err_pid10352.log
#
# If you would like to submit a bug report, please visit:
# https://bugs.launchpad.net/ubuntu/+source/openjdk-6/
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#
Aborted (core dumped)

Revision history for this message
branflake2267 (branflake2267) wrote :

I just updated to ubuntu 8.04 hardy. Went to load my eclipse, and I got this. I have tried sun-Java-6 and sun-Java-1.5 and gcj-java all with the same error.

This bug really sucks. Has anybody found a workaround to get eclipse running.

#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007f5b68dd18f1, pid=18327, tid=1079228752
#
# Java VM: OpenJDK 64-Bit Server VM (1.6.0-b09 mixed mode linux-amd64)
# Problematic frame:
# C [libxcb.so.1+0x98f1]
#
# An error report file with more information is saved as:
# /home/branflake2267/hs_err_pid18327.log
#
# If you would like to submit a bug report, please visit:
# https://bugs.launchpad.net/ubuntu/+source/openjdk-6/
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.

Revision history for this message
Martin Holtz (martin-holtz) wrote :

Hi,
eclipse crashes while using svn update

#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00002b8c63b71fa0, pid=30903, tid=1082161488
#
# Java VM: OpenJDK 64-Bit Server VM (1.6.0-b09 mixed mode linux-amd64)
# Problematic frame:
# V [libjvm.so+0x20afa0]
#
# An error report file with more information is saved as:
# /tmp/hs_err_pid30903.log
#
# If you would like to submit a bug report, please visit:
# https://bugs.launchpad.net/ubuntu/+source/openjdk-6/
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.

Changed in openjdk-6:
status: Incomplete → Confirmed
Revision history for this message
Martin Holtz (martin-holtz) wrote :

have an look there:
http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6659207

"A DESCRIPTION OF THE PROBLEM :
  Program consistently fails in 64-bit jvm (client & server) but works in 32-bit jvm. Get access violation in hotspot compiler thread (but reported as "outside the jvm in native code" - I think incorrectly)."

Revision history for this message
Matthias Klose (doko) wrote :

please recheck with openjdk-6 in intrepid, or openjdk-6 6b09dfsg-0ubuntu2~ppa1 in the openjdk PPA.

Changed in openjdk-6:
status: Confirmed → Incomplete
Revision history for this message
neiltait (neil-tait) wrote :

no joy with openjdk-6...

Revision history for this message
nyarnon (cabal) wrote :

Here's my 50cents:

#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0xb6c04186, pid=6663, tid=3025165200
#
# Java VM: OpenJDK Client VM (1.6.0-b09 mixed mode, sharing linux-x86)
# Problematic frame:
# V [libjvm.so+0x31d186]
#
# An error report file with more information is saved as:
# /home/roelf/hs_err_pid6663.log

Revision history for this message
Adrian Custer (acuster) wrote :

Problem perrsists with:
 OpenJDK package 6b10-0ubuntu1~ppa3

/usr/lib/jvm/java-6-openjdk/bin/java -version
  java version "1.6.0"
  OpenJDK Runtime Environment (build 1.6.0-b10)
  OpenJDK 64-Bit Server VM (build 1.6.0-b10, mixed mode)

and Ganymede RC2
  Version: 3.4.0
  Build id: I20080523-0100

works fine with java 1.5.0
java -version
  java version "1.5.0_15"
  Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_15-b04)
  Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_15-b04, mixed mode)

--adrian

Revision history for this message
Alexey Dubrovskiy (alexey-dubrovskiy) wrote :

I had the same problem. Then I turned off 'Build Automatically'. And it works without crashing!

Revision history for this message
Simon (simon-beverton) wrote :

Thanks guys for the suggestion to copy in the libjvm.so from jdk-7- seems to have done the job for now.
I am using CDT 5 inside eclipse Ganymede with about 6 projects and as of yesterday one project decided to crash the system with the unexpected error in libjvm.so. I think it was the action of the C++ indexer kicking in that would eventually crash the system reproduceably.

Revision history for this message
Amos Brocco (mriya3) wrote :

This bug should be fixed with jdk 1.6u10 b28 (download from http://download.java.net/jdk6/binaries/ , file called jdk-6u10-rc-bin-b28-linux-amd64-21_jul_2008.bin).

You can create a package using a patched version of java-package (I've made one you can download from http://diuf.unifr.ch/pai/people/broccoa/jpkg/ ... install the deb and then run make-jpkg jdk-6u10-rc-bin-b28-linux-amd64-21_jul_2008.bin , install the resulting deb file, and run update-alternatives --config java )

I'm working with Eclipse 3.3.2 and it seems stable (previously I had to use IBM JVM because of this bug)

Revision history for this message
gnuboi (chunghaulee) wrote :

Thank you, Amos. I'm running 8.04 and your post was the best way to get an updated Java. My Eclipse seems much more stable now.

Revision history for this message
joe (mjoachimiak) wrote :

I have a little bit similar problems with eclipse 3.4 when working with openjdk 1.6.0.
When Eclipse starts and I press New->Project->C++ it crashes:

# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007fad9d3ebbc4, pid=7054, tid=1108134224
#
# Java VM: OpenJDK 64-Bit Server VM (1.6.0_0-b11 mixed mode linux-amd64)
# Problematic frame:
# V [libjvm.so+0x227bc4]
#
# An error report file with more information is saved as:
# /home/peloton/programs/eclipse/hs_err_pid7054.log
#
# If you would like to submit a bug report, please visit:
# https://bugs.launchpad.net/ubuntu/+source/openjdk-6/
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.

When I use Sun Java(TM) Runtime Environment (JRE) 6 it goes without error.
I will wait for jdk 1.6u10 b28. I hope it will fix also my problem.

Matthias Klose (doko)
tags: added: eclipse local-binary swt-gtk
Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for openjdk-6 (Ubuntu) because there has been no activity for 60 days.]

Changed in openjdk-6 (Ubuntu):
status: Incomplete → Expired
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.