[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#37671: 27.0.50; Segmentation fault with --fg-daemon on Linux
From: |
Eli Zaretskii |
Subject: |
bug#37671: 27.0.50; Segmentation fault with --fg-daemon on Linux |
Date: |
Thu, 10 Oct 2019 14:03:57 +0300 |
> From: Frank Terbeck <ft@bewatermyfriend.org>
> Date: Wed, 09 Oct 2019 03:55:23 +0200
>
>
> I am running emacs in Linux. The base system is an up-to-date Debian
> sid. Right now I am starting emacs via systemd and emacs' --fg-daemon
> flag. I am mostly attaching GUI clients to this server instance. The
> graphical user interface is a non-reparenting window manager (namely
> xmonad). This build uses --with-x-toolkit=lucid --with-cairo as well as
> --with-libsystemd.
>
> The issue is emacs crashing, plainly. And I have not been able to iso-
> late and specifics that are required for crashes to happen. I've seen
> this sporadically with different builds of emacs from its master branch.
> To look into it more closely, I rebuilt emacs with debugging symbols.
> And luckily, I didn't have to wait very long. Emacs crashed within a
> day.
>
> I didn't so anything spectacular either. I think I was just switching
> from a buffer to another. And emacs' uptime was pretty low too. Like an
> hour or two, if I'd have to wager a guess.
>
> Here is the stack-trace, as recorded by systemd's coredumpctl:
>
> Stack trace of thread 1679:
> #0 0x00007f34412112a7 __pthread_kill (libpthread.so.0)
> #1 0x000055d90de0f632 deliver_process_signal (emacs-27.0.50)
> #2 0x000055d90df7be10 deliver_child_signal (emacs-27.0.50)
> #3 0x00007f3441214510 __restore_rt (libpthread.so.0)
> #4 0x00007f3440f92d2f __GI___poll (libc.so.6)
> #5 0x00007f3441e3809e n/a (libglib-2.0.so.0)
> #6 0x00007f3441e38403 g_main_loop_run (libglib-2.0.so.0)
> #7 0x00007f34420a3726 n/a (libgio-2.0.so.0)
> #8 0x00007f3441e60d0d n/a (libglib-2.0.so.0)
> #9 0x00007f3441209fb7 start_thread (libpthread.so.0)
> #10 0x00007f3440f9d2ef __clone (libc.so.6)
Thanks.
Looks like the crash was in Glib?
It is strange that you see a problem that no one else sees, and you
say you have had it for quite some time.
Can you run Emacs under GDB? I hope it will show a better backtrace
if you do so.