java assert failure: *** Error in `/usr/lib/jvm/java-7-openjdk-amd64/bin/java': double free or corruption (!prev): 0x00007f46303fa9a0 ***

[Lonnie Lee Best]

openjdk-7 often crashes when I launch freemind:
sudo apt-get install freemind

This bug report is an example of one of these crashes.

ProblemType: Crash
DistroRelease: Ubuntu 15.04
Package: openjdk-7-jre-headless 7u79-2.5.5-0ubuntu1
ProcVersionSignature: Ubuntu 3.19.0-14.14-generic 3.19.3
Uname: Linux 3.19.0-14-generic x86_64
ApportVersion: 2.17.1-0ubuntu2
Architecture: amd64
AssertionMessage: *** Error in `/usr/lib/jvm/java-7-openjdk-amd64/bin/java': double free or corruption (!prev): 0x00007f46303fa9a0 ***
CurrentDesktop: Unity
Date: Fri Apr 17 12:58:07 2015
ExecutablePath: /usr/lib/jvm/java-7-openjdk-amd64/jre/bin/java
InstallationDate: Installed on 2015-04-17 (0 days ago)
InstallationMedia: Ubuntu 15.04 "Vivid Vervet" - Beta amd64 (20150417)
ProcCmdline: /usr/lib/jvm/java-7-openjdk-amd64/bin/java -Xmx256M -Dgnu.java.awt.peer.gtk.Graphics=Graphics2D -Dfreemind.base.dir=/usr/share/freemind -cp ::/usr/share/freemind/lib/freemind.jar:/usr/share/java/SimplyHTML.jar:/usr/share/java/gnu-regexp.jar:/usr/share/java/jibx-run-1.1.6a.jar:/usr/share/java/xpp3.jar:/usr/share/freemind/lib/bindings.jar:/usr/share/java/forms.jar:/usr/share/freemind freemind.main.FreeMindStarter
ProcEnviron:
 LANGUAGE=en_US
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=en_US.UTF-8
 SHELL=/bin/bash
Signal: 6
SourcePackage: openjdk-7
StacktraceTop:
 __libc_message (do_abort=do_abort@entry=1, fmt=fmt@entry=0x7f463820a1a8 "*** Error in `%s': %s: 0x%s ***\n") at ../sysdeps/posix/libc_fatal.c:175
 malloc_printerr (ptr=<optimized out>, str=0x7f463820a2d8 "double free or corruption (!prev)", action=1) at malloc.c:4965
 _int_free (av=<optimized out>, p=<optimized out>, have_lock=0) at malloc.c:3834
 __GI___libc_free (mem=<optimized out>) at malloc.c:2950
 _XlcDestroyLocaleDataBase () from /usr/lib/x86_64-linux-gnu/libX11.so.6
Title: java assert failure: *** Error in `/usr/lib/jvm/java-7-openjdk-amd64/bin/java': double free or corruption (!prev): 0x00007f46303fa9a0 ***
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip libvirtd lpadmin plugdev sambashare sudo

[Lonnie Lee Best (launchpad-startport)]

I've experienced this crash in both 14.04 and 15.04.

[Apport retracing service (apport)]

StacktraceTop:
 __libc_message (do_abort=do_abort@entry=1, fmt=fmt@entry=0x7f463820a1a8 "*** Error in `%s': %s: 0x%s ***\n") at ../sysdeps/posix/libc_fatal.c:175
 malloc_printerr (ptr=<optimized out>, str=0x7f463820a2d8 "double free or corruption (!prev)", action=1) at malloc.c:4965
 _int_free (av=<optimized out>, p=<optimized out>, have_lock=0) at malloc.c:3834
 __GI___libc_free (mem=<optimized out>) at malloc.c:2950
 _XlcDestroyLocaleDataBase (lcd=lcd@entry=0x7f463037b6b0) at ../../../src/xlibi18n/lcDB.c:1252

[Lonnie Lee Best (launchpad-startport)]

This bug has gotten worse in Ubuntu 15.04

More often than not, I cannot successfully launch Freemind without having to kill the java process several times.

It goes like this:

1) Try to launch freemind
2) Freemind locks up while trying to launch
3) Kill the java process that is locked up
4) Go back to step 1.

I have to repeat these steps 3 to 5 times before I can successfully launch freemind.

[Lonnie Lee Best (launchpad-startport)]

Today, I had to kill the java ( openjdk-7 ) process 4 times before freemind would launch.

Can someone please confirm this bug?

[Tiago Stürmer Daitx (tdaitx)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. Unfortunately, we have been unable to reproduce the error after following the steps provided. We were able to start and run FreeMind without errors multiple times. Also, there are no reports of similar instabilities in the FreeMind bug tracking tool.

As your original install media was a Beta release, please make sure you are running the latest package versions available. Let us know if you system is stable overall and if only this package is being affected.

In order to help tracking down this issue, please set the MALLOC_CHECK_ environment variable to 2 before calling freemind. Run the following in a terminal:
$ export MALLOC_CHECK_=2
$ freemind
as described in http://stackoverflow.com/a/2903152

You might also try running freemind through valgrind:
$ valgrind freemind
and report back with the full console ouput.

Thanks!

[Lonnie Lee Best (launchpad-startport)]

I've reinstall 15.10 since the beta. Here is the output of the command requested:

~$ export MALLOC_CHECK_=2
~$ freemind
Checking Java Version...
java: Fatal IO error 2 (No such file or directory) on X server :0.
~$ ^C
~$

[Lonnie Lee Best (launchpad-startport)]

valgrind freemind
==14509== Memcheck, a memory error detector
==14509== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al.
==14509== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright info
==14509== Command: /usr/bin/freemind
==14509==
==14510== Warning: invalid file descriptor -1 in syscall close()
==14511==
==14511== HEAP SUMMARY:
==14511== in use at exit: 11,290 bytes in 80 blocks
==14511== total heap usage: 82 allocs, 2 frees, 11,338 bytes allocated
==14511==
==14511== LEAK SUMMARY:
==14511== definitely lost: 0 bytes in 0 blocks
==14511== indirectly lost: 0 bytes in 0 blocks
==14511== possibly lost: 0 bytes in 0 blocks
==14511== still reachable: 11,290 bytes in 80 blocks
==14511== suppressed: 0 bytes in 0 blocks
==14511== Rerun with --leak-check=full to see details of leaked memory
==14511==
==14511== For counts of detected and suppressed errors, rerun with: -v
==14511== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
==14510==
==14510== HEAP SUMMARY:
==14510== in use at exit: 3,126 bytes in 80 blocks
==14510== total heap usage: 81 allocs, 1 frees, 3,158 bytes allocated
==14510==
==14510== LEAK SUMMARY:
==14510== definitely lost: 0 bytes in 0 blocks
==14510== indirectly lost: 0 bytes in 0 blocks
==14510== possibly lost: 0 bytes in 0 blocks
==14510== still reachable: 3,126 bytes in 80 blocks
==14510== suppressed: 0 bytes in 0 blocks
==14510== Rerun with --leak-check=full to see details of leaked memory
==14510==
==14510== For counts of detected and suppressed errors, rerun with: -v
==14510== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
==14520== Warning: invalid file descriptor -1 in syscall close()
==14520==
==14520== HEAP SUMMARY:
==14520== in use at exit: 19,538 bytes in 133 blocks
==14520== total heap usage: 240 allocs, 107 frees, 72,212 bytes allocated
==14520==
==14520== LEAK SUMMARY:
==14520== definitely lost: 0 bytes in 0 blocks
==14520== indirectly lost: 0 bytes in 0 blocks
==14520== possibly lost: 0 bytes in 0 blocks
==14520== still reachable: 19,538 bytes in 133 blocks
==14520== suppressed: 0 bytes in 0 blocks
==14520== Rerun with --leak-check=full to see details of leaked memory
==14520==
==14520== For counts of detected and suppressed errors, rerun with: -v
==14520== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
==14509== Warning: invalid file descriptor -1 in syscall close()
==14523==
==14523== HEAP SUMMARY:
==14523== in use at exit: 48,610 bytes in 168 blocks
==14523== total heap usage: 360 allocs, 192 frees, 141,916 bytes allocated
==14523==
==14523== LEAK SUMMARY:
==14523== definitely lost: 0 bytes in 0 blocks
==14523== indirectly lost: 0 bytes in 0 blocks
==14523== possibly lost: 0 bytes in 0 blocks
==14523== still reachable: 48,610 bytes in 168 blocks
==14523== suppressed: 0 bytes in 0 blocks
==14523== Rerun with --leak-check=full to see details of leaked memory
==14523==
==14523== For counts of detected and suppressed errors, rerun with: -v
==14523== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
==14525==
=...

[Lonnie Lee Best (launchpad-startport)]

This remains an issue for me in Ubuntu 15.10. Most of the time, when I launch freemind, the launching-progress-bar only makes it to about 10% as can be seen in this screenshot:
http://neartalk.com/ss/2015-12-30_001_300x200.png

Sometimes I have to kill the dependent java process 4 or 5 times before freemind will launch successfully.

Above I've provided more details about this issue, and I'm hoping someone who understands those details can provide me a suggestion.

[Lonnie Lee Best (launchpad-startport)]

Here's a screen shot of the System Monitor during one of these lockups:
http://neartalk.com/ss/2015-12-30_002_1528x523.png

[Lonnie Lee Best (launchpad-startport)]

Today, I'm unable to run freemind due to this bug; now killing the java process has no effect next launch.

As an alternative, I tried to use freeplane instead of freemind:
sudo apt-get install freeplane freemind

Both of these application lock up due to this bug. In freeplane, I can get into the GUI, but as soon as I try to navigate the file system (before getting to a mindmap file to open) it locks up too. With Freemind, I can't even get into the GUI.

[Lonnie Lee Best (launchpad-startport)]

Freeplane seem to have the same dependencies as freemind (see screenshot I posted earlier for freemind):
http://neartalk.com/ss/2016-01-12_001_1204x314.png

[Lonnie Lee Best (launchpad-startport)]

Sorry, I meant to post this:
http://neartalk.com/ss/2016-01-12_002_1189x190.png

[Lonnie Lee Best (launchpad-startport)]

Can someone please at least offer some advise toward a workaround that may fix this issue?

I've spend countless hour making a freemind map that has information that I must reference frequently. In order to view this info now, I'm having to remote into a Windows OS with freemind installed, because I'm no longer able to open freemind on Ubuntu Linux (where the map was built).

[Lonnie Lee Best (launchpad-startport)]

I'm trying to get help here:
http://unix.stackexchange.com/questions/257881/cant-launch-freemind-in-ubuntu-15-10-please-advise

[Tiago Stürmer Daitx (tdaitx)]

Hi Lonnie,

Could you please check if the workaround bellow works for you? From
https://bugs.launchpad.net/ubuntu/+source/openjdk-8/+bug/1510009/comments/24
(there is also a report from
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=798131 saying this
might fix the issue).

"I confirm the following workaround is working: comment out the line
'assistive_technologies=org.GNOME.Accessibility.AtkWrapper' in
/etc/java-7-openjdk/accessibility.properties

I had Freeplane systematically hanging when browsing or even saving a
file, and the works around the problem perfectly. But this should
definitely be fixed as many Java apps seems to be affected...."

On Tue, Jan 26, 2016 at 9:56 PM, Lonnie Lee Best
<email address hidden> wrote:
> I'm trying to get help here:
> http://unix.stackexchange.com/questions/257881/cant-launch-freemind-in-ubuntu-15-10-please-advise
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1445598
>
> Title:
> java assert failure: *** Error in `/usr/lib/jvm/java-7-openjdk-
> amd64/bin/java': double free or corruption (!prev): 0x00007f46303fa9a0
> ***
>
> Status in FreeMind:
> Incomplete
> Status in freeplane:
> New
> Status in java-common:
> New
> Status in Light Display Manager:
> New
> Status in systemd:
> New
> Status in openjdk-7 package in Ubuntu:
> Incomplete
>
> Bug description:
> openjdk-7 often crashes when I launch freemind:
> sudo apt-get install freemind
>
> This bug report is an example of one of these crashes.
>
> ProblemType: Crash
> DistroRelease: Ubuntu 15.04
> Package: openjdk-7-jre-headless 7u79-2.5.5-0ubuntu1
> ProcVersionSignature: Ubuntu 3.19.0-14.14-generic 3.19.3
> Uname: Linux 3.19.0-14-generic x86_64
> ApportVersion: 2.17.1-0ubuntu2
> Architecture: amd64
> AssertionMessage: *** Error in `/usr/lib/jvm/java-7-openjdk-amd64/bin/java': double free or corruption (!prev): 0x00007f46303fa9a0 ***
> CurrentDesktop: Unity
> Date: Fri Apr 17 12:58:07 2015
> ExecutablePath: /usr/lib/jvm/java-7-openjdk-amd64/jre/bin/java
> InstallationDate: Installed on 2015-04-17 (0 days ago)
> InstallationMedia: Ubuntu 15.04 "Vivid Vervet" - Beta amd64 (20150417)
> ProcCmdline: /usr/lib/jvm/java-7-openjdk-amd64/bin/java -Xmx256M -Dgnu.java.awt.peer.gtk.Graphics=Graphics2D -Dfreemind.base.dir=/usr/share/freemind -cp ::/usr/share/freemind/lib/freemind.jar:/usr/share/java/SimplyHTML.jar:/usr/share/java/gnu-regexp.jar:/usr/share/java/jibx-run-1.1.6a.jar:/usr/share/java/xpp3.jar:/usr/share/freemind/lib/bindings.jar:/usr/share/java/forms.jar:/usr/share/freemind freemind.main.FreeMindStarter
> ProcEnviron:
> LANGUAGE=en_US
> PATH=(custom, no user)
> XDG_RUNTIME_DIR=<set>
> LANG=en_US.UTF-8
> SHELL=/bin/bash
> Signal: 6
> SourcePackage: openjdk-7
> StacktraceTop:
> __libc_message (do_abort=do_abort@entry=1, fmt=fmt@entry=0x7f463820a1a8 "*** Error in `%s': %s: 0x%s ***\n") at ../sysdeps/posix/libc_fatal.c:175
> malloc_printerr (ptr=<optimized out>, str=0x7f463820a2d8 "double free or corruption (!prev)", action=1) at malloc.c:4965
> _int_free (av=<optimized out>, p=<optimized o...

[Lonnie Lee Best (launchpad-startport)]

Tiago, thank you so much for the workaround!

I've been battling this issue since April of 2015, and I confirm that the workaround you've provided works for me in Ubuntu 15.10.

With the workaround you've referenced, the person assigned to this bug should be able to finally make some progress on getting this solved for everyone.

Thanks again!

[Thorsten Glaser (mirabilos)]

Cannot reproduce with 8u392-ga-1~deb9u1 so I presume this fixed in Ubuntu as well, since it uses the same packages these days.