View previous topic :: View next topic |
Author |
Message |
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Thu Oct 09, 2003 7:56 pm Post subject: evolution segmentation fault on start |
|
|
Hi,
Since I emerged evolution 1.4.5 a couple of weeks ago I have been unable to start the application.
I thought that maybe it was a bad ebuild or something, so I waited patiently and ran Code: | emerge -u world --deep | hoping that it would magically be fixed.
I also did a two or three times, without getting my problems solved. However the first time it broke I got ?-marks instead of icons in the splash screen, not I got the right icons back.
When I launch from the icon I just get a stupid popup after the splash with a segmentation fault.
Launching from the command line gives me this output (and the popup):
Code: |
enoch ~ $> evolution
(evolution:24104): shortcut-bar-CRITICAL **: file e-shortcut-model.c: line 288 (e_shortcut_model_real_add_group): assertion `group_name != NULL' failed
(evolution:24104): shortcut-bar-CRITICAL **: file e-shortcut-model.c: line 340 (e_shortcut_model_add_item): assertion `group_num < shortcut_model->groups->len' failed
(evolution:24104): shortcut-bar-CRITICAL **: file e-shortcut-model.c: line 340 (e_shortcut_model_add_item): assertion `group_num < shortcut_model->groups->len' failed
(evolution:24104): shortcut-bar-CRITICAL **: file e-shortcut-model.c: line 340 (e_shortcut_model_add_item): assertion `group_num < shortcut_model->groups->len' failed
(evolution:24104): shortcut-bar-CRITICAL **: file e-shortcut-model.c: line 340 (e_shortcut_model_add_item): assertion `group_num < shortcut_model->groups->len' failed
(evolution:24104): shortcut-bar-CRITICAL **: file e-shortcut-model.c: line 340 (e_shortcut_model_add_item): assertion `group_num < shortcut_model->groups->len' failed
(evolution:24104): shortcut-bar-CRITICAL **: file e-shortcut-bar.c: line 717 (e_shortcut_bar_set_view_type): assertion `group_num < shortcut_bar->groups->len' failed
(evolution:24104): e-table-CRITICAL **: file e-tree-memory.c: line 95 (e_tree_memory_path_depth): assertion `path != NULL' failed
|
I am at my wits end. I hope somebody can give me a hint.
Thanks
/Lazlo |
|
Back to top |
|
|
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Thu Oct 09, 2003 11:24 pm Post subject: no luck with 1.4.4 either |
|
|
Just tried
Code: |
emerge -C evolution
emerge /usr/portage/net-mail/evolution/evolution-1.4.4.ebuild
|
Still same error message.
Tried Code: | evolution --force-shutdown | without luck
I don't recall having success running it as root before. Now I got the setup wizard. Once I completed it I got the same segment fault as always.
Here's my settings, running Gnome 2.4 on a Athlon XP 2000 with 512mb ram and 160gb disc space.
Code: |
CHOST="i686-pc-linux-gnu"
CFLAGS="-mcpu=athlon-xp -O3 -fomit-frame-pointer -funroll-loops -pipe"
USE="X gtk2 gnome -qt"
ACCEPT_KEYWORDS="~x86"
|
Going to try upgrading again to see what that'll bring me.
Can't help having a feeling of going in circles though ... :(
[/code] |
|
Back to top |
|
|
piefke n00b
Joined: 05 Sep 2003 Posts: 57
|
Posted: Thu Oct 09, 2003 11:35 pm Post subject: |
|
|
have you tried to downgrade your cflags?
Code: | CFLAGS="-mcpu=athlon-xp -O2 -fomit-frame-pointer -pipe" |
|
|
Back to top |
|
|
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Fri Oct 10, 2003 7:46 am Post subject: |
|
|
Tried with your flags.
Still no luck. Same error.
Is there any dependencies to evolution I should try to reemerge ?
Any and all ideas are welcome!
TIA
/Lazlo |
|
Back to top |
|
|
russo79 n00b
Joined: 22 May 2003 Posts: 27
|
Posted: Fri Oct 10, 2003 8:27 am Post subject: |
|
|
Did you tried to do a revdep-rebuild???
If there is some problem with the libs, they will probably be fixed after that.
One other thing, if you use gnome-2.4, evolution crashes after some time of using it.
There's a bug report for that. |
|
Back to top |
|
|
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Fri Oct 10, 2003 4:36 pm Post subject: |
|
|
I tried it and got:
Code: | enoch root # revdep-rebuild evolution |
It then exited first with a complaint about gaim. Which I unmerged.
Then x11/xsod which I then upgraded from 0.7 to 2.2.5 (strange that this wasn't caught by a emerge -u world --deep ?)
Then I ran the command again and got:
Code: | enoch root # revdep-rebuild evolution
Checking reverse dependencies...
Packages containing binaries and libraries broken by any package update,
will be recompiled.
Collecting system binaries and libraries... done.
(/root/.revdep-rebuild.1_files)
Collecting complete LD_LIBRARY_PATH... done.
(/root/.revdep-rebuild.2_ldpath)
Checking dynamic linking consistency...
done.
(/root/.revdep-rebuild.3_rebuild)
Assigning files to ebuilds... Nothing to rebuild
Evaluating package order... done.
(/root/.revdep-rebuild.5_order)
|
That didn't do anything, so obviously it didn't solve my problem.
I tried to emerge evolution then. Still no luck.
Thanks for the help so far.
I don't know what to try next.
/Lazlo |
|
Back to top |
|
|
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Mon Oct 13, 2003 10:07 am Post subject: |
|
|
Was I doing the revdep-rebuild correctly ???
I just discovered/remembered that my make.conf file was accidentially overwritten in etc-update (user error). So I got my [~x86] flag removed.
This might have caused some packages to be accidentally downgraded.
Which might again have an impact on evolution ??
So how can I learn witch packages evolution depends on, so I can try to remerge them ?
Maybe someone with a working gnome 2.4 and evolution could post a list of his/her installed packages and their versions.
/Lazlo |
|
Back to top |
|
|
johpe916 n00b
Joined: 13 Sep 2002 Posts: 45 Location: Sweden
|
Posted: Mon Oct 13, 2003 12:03 pm Post subject: |
|
|
Try using gdb. It is a debugger. "emerge -s gdb"
It is very handy when dealing with segmentation faults.
gdb evolution
and at the gdb prompt type "run"
It might give som extra information. Especially if the segmentation fault isn´t caused by evolution, but some underlaying package.
/Johan |
|
Back to top |
|
|
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Sun Oct 19, 2003 7:21 pm Post subject: |
|
|
That seemed to get me a little bit further.
Here's what I did and what happened:
gdb evolution
gave me some error in libgal
I had no idea what that meant, so I tried brute force:
cd /usr/portage
grep -R libgal *
and found out that there is a package called gnome-extras/gal
I (r)emerge that and that helped a little bit. I now get beyond the splashscreen and get the main window up for a splitsecond. Then it crashes once again.
This time gdb reports some broken pipes from libc which I guessed that a remerge of glibc might cure.
It did, anyways I didn't get that error again. But I got this:
Code: |
(no debugging symbols found)...The program 'evolution' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadPixmap (invalid Pixmap parameter)'.
(Details: serial 1878 error_code 4 request_code 54 minor_code 0)
(Note to programmers: normally, X errors are reported asynchronously;
that is, you will receive the error a while after causing it.
To debug your program, run it with the --sync command line
option to change this behavior. You can then get a meaningful
backtrace from your debugger if you break on the gdk_x_error() function.)
The program 'evolution' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadPixmap (invalid Pixmap parameter)'.
(Details: serial 1884 error_code 4 request_code 54 minor_code 0)
(Note to programmers: normally, X errors are reported asynchronously;
that is, you will receive the error a while after causing it.
To debug your program, run it with the --sync command line
option to change this behavior. You can then get a meaningful
backtrace from your debugger if you break on the gdk_x_error() function.)
The program 'evolution' received an X Window System error.
This probably reflects a bug in the program.
The error was 'RenderBadPicture (invalid Picture parameter)'.
(Details: serial 1885 error_code 177 request_code 156 minor_code 7)
(Note to programmers: normally, X errors are reported asynchronously;
that is, you will receive the error a while after causing it.
To debug your program, run it with the --sync command line
option to change this behavior. You can then get a meaningful
backtrace from your debugger if you break on the gdk_x_error() function.)
|
Or course I tried to break gdk_x_error in gdb as the error msg requests. But just got the message that that function was not defined...
It didn't solve my problem. But thanks for teaching me if not to fish, then at least how to get the line in the water :)
/Lasse |
|
Back to top |
|
|
liquidx Retired Dev
Joined: 21 Aug 2002 Posts: 56 Location: London, UK
|
Posted: Sun Oct 19, 2003 9:44 pm Post subject: |
|
|
judging from the problems you have .. it might be best to downgrade those CFLAGS and try merging all the deps from evolution, maybe using --deep.
i've seen -O3 produce some weird code with gcc-3.3.1 on pentium2. |
|
Back to top |
|
|
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Sun Oct 19, 2003 9:49 pm Post subject: |
|
|
Oh.
But that was the first thing I did.
Now I use:
CFLAGS="-mcpu=athlon-xp -Os -fomit-frame-pointer -pipe"
Which I belive is on the conservative side ?
Can you explain to me how I merge all deps from evolution ?
I don't know what they are.
/Lasse |
|
Back to top |
|
|
Lazlo n00b
Joined: 25 Jan 2003 Posts: 66 Location: Gl. Lejre, Denmark
|
Posted: Sun Oct 26, 2003 9:16 am Post subject: evolution segmentation fault on start [solved] |
|
|
Finally someone took the time to answer my question. Thanks to the nice guys at the #gentoo irc channel on freenode.
The trick to remerge evolution with all back dependencies was:
Code: | emerge -eu evolution |
It remerged 141 packages and took all night, but it was worth it. Evolution finally works again.
Next time I run into something that isn't solved after more than a month, I think I'll go back to windows for good. So lets not hope that will happen...
/Lazlo |
|
Back to top |
|
|
rincewind Apprentice
Joined: 16 Aug 2002 Posts: 172 Location: Essen, Germany
|
Posted: Sun Oct 26, 2003 10:31 am Post subject: |
|
|
But Windows also isnt solved yet for more than 10 years ..... |
|
Back to top |
|
|
|