Bug 1197665

Created attachment 997019 [details]
File: backtrace

Created attachment 997020 [details]
File: cgroup

Created attachment 997021 [details]
File: core_backtrace

Created attachment 997022 [details]
File: dso_list

Created attachment 997023 [details]
File: environ

Created attachment 997024 [details]
File: limits

Created attachment 997025 [details]
File: maps

Created attachment 997026 [details]
File: open_fds

Created attachment 997027 [details]
File: proc_pid_status

Created attachment 997028 [details]
File: var_log_messages

Another user experienced a similar problem:

Trying to generate backtrace from the running Evolution.

Something like

gdb -nx --batch -ex 'thread apply all backtrace' --pid=$(pidof evolution)|&tee evolution-backtrace.txt

reporter:       libreport-2.1.11
backtrace_rating: 4
cmdline:        gdb -batch -iex 'add-auto-load-safe-path /var/cache/abrt-di/usr/lib/debug' -iex 'add-auto-load-scripts-directory /var/cache/abrt-di/usr/lib/debug' -ex 'set debug-file-directory /usr/lib/debug:/var/cache/abrt-di/usr/lib/debug' -ex 'file /usr/bin/nautilus' -ex 'core-file ./coredump' -ex 'thread apply all backtrace 1024 full' -ex 'info sharedlib' -ex 'print (char*)__abort_msg' -ex 'print (char*)__glib_assert_msg' -ex 'info all-registers' -ex disassemble
crash_function: dump_core
executable:     /usr/bin/gdb
kernel:         3.10.0-229.el7.x86_64
package:        gdb-7.6.1-64.el7
reason:         gdb killed by SIGABRT
runlevel:       N 5
type:           CCpp
uid:            1000

gdb-upstream-framefilters-1of2.patch:
+apply_frame_filter():
+  struct gdbarch *gdbarch = NULL;
+  cleanups = ensure_python_env (gdbarch, current_language);
+  TRY_CATCH (except, RETURN_MASK_ALL)
+    {
+      gdbarch = get_frame_arch (frame);
+    }

This really cannot work, ensure_python_env is called with gdbarch=NULL.

For some reason initial upstream commit already has it right:
1e611234ee3f4a1d2434f3fe7530cab87c936e0d

This buggy backport comes from:
commit 89fbbfccaf093284eadb1af638486ee5b1701a81
Author: Jan Kratochvil <jan.kratochvil>
Date:   Tue May 21 13:54:33 2013 +0200
+# Backported Python frame filters (Phil Muldoon).
+Patch836: gdb-upstream-framefilters-1of2.patch
+* Tue May 21 2013 Jan Kratochvil <jan.kratochvil> - 7.6-30.fc19
+- Backported Python frame filters (Phil Muldoon).

I actually see the same error on 1e611234ee3f4a1d2434f3fe7530cab87c936e0d.  This has been fixed by:

commit 21909fa1c6d934bfa0c7ad3ef95909db48f6f756
Author: Tom Tromey <tromey>
Date:   Wed Jan 22 08:10:01 2014 -0700

So the right fix would be to backport this patch.

Created attachment 1006042 [details]
output of gdb --pid=$(pidof evolution)

Unfortunately, gdb-7.6.1-65.el7.x86_64 doesn't fix the issue, or at least I have still Python exceptions in the output.

I don't see GDB crashing anymore on your output.  This exception happens when GDB cannot locate a frame block refered by the backtrace (GDB then raises a Python exception), but the original problem has been fixed by the backported patch.

FTR: I provided a scratch build for Matej.

Matej, can you confirm that the original problem has been fixed?  I intend to check the patch in in the next few days.

Created attachment 1025041 [details]
output of gdb

(In reply to Sergio Durigan Junior from comment #17)
> Matej, can you confirm that the original problem has been fixed?  I intend
> to check the patch in in the next few days.

Is this what you want? I got it with gdb-7.6.1-65.el7.x86_64

*** Bug 1221133 has been marked as a duplicate of this bug. ***

I've updated from release 64 to gdb-7.6.1-68.el7.x86_64 but I still can reproduce the crash. Is it supposed to be fixed in any of these versions?

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://rhn.redhat.com/errata/RHBA-2015-2089.html