Bug 1513807 - xorg-x11-server quit unexpectedly Xwayland killed by SIGABRT
Summary: xorg-x11-server quit unexpectedly Xwayland killed by SIGABRT
Keywords:
Status: CLOSED EOL
Alias: None
Product: Fedora
Classification: Fedora
Component: mutter
Version: 27
Hardware: x86_64
OS: Linux
unspecified
urgent
Target Milestone: ---
Assignee: Florian Müllner
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2017-11-16 01:57 UTC by Daniel Rowe
Modified: 2018-11-30 19:35 UTC (History)
25 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2018-11-30 19:35:34 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
logs (43.43 KB, text/plain)
2017-11-16 01:57 UTC, Daniel Rowe
no flags Details
backtrace captured with gdb (45.28 KB, text/plain)
2017-11-16 11:01 UTC, Daniel Rowe
no flags Details

Description Daniel Rowe 2017-11-16 01:57:15 UTC
Created attachment 1353152 [details]
logs

Description of problem:

Under way land as soon as display wake form sleep:

gnome-shell[1528]: segfault at 38 ip 00007efc97ccd2e0 sp 00007ffd5c69d688 error 4 in libmutter-1.so.0.0.0[7efc97c67000+16b000]



Version-Release number of selected component (if applicable):

All latest Fedora 27 Workstation/

How reproducible:

Every time

Steps to Reproduce:
1. Log in
2. Let display sleep
3. Wake display

Actual results:

Crash

Expected results:

Works as it did in Fedora 26

Additional info:

https://retrace.fedoraproject.org/faf/reports/bthash/18d455894df7d9531ace5a146f495863b59721d1

Comment 1 Olivier Fourdan 2017-11-16 07:36:23 UTC
That's gnome-shell segfaulting, Xwayland loses its connection to the Wayland compositor and aborts.

So the problem lies in gnome-shell, Xwayland is the "victim" here, like any other Wayland client.

Can you capture a backtrace of gnome-shell (with the symbols)?

Comment 2 Olivier Fourdan 2017-11-16 07:37:05 UTC
Meanwhile, moving to mutter, the crash in gnome-shell is most likely a mutter bug.

Comment 3 Daniel Rowe 2017-11-16 09:06:40 UTC
OK are there any docs to capture a backtrace of with the symbols? This happens every time so will be easy to trigger.

Comment 4 Olivier Fourdan 2017-11-16 10:16:16 UTC
First step would be to install the debuginfo packages for gnome-shell ("dnf debuginfo-install gnome-shell" as root).

Then you can check if "abrt-cli ls" lists the crash for gnome-shell, in which case you could try to upload the report with "abrt-cli report <directory>", that would be the easiest way, I reckon.

Failing that, it would require attaching gdb to the running gnome-shell process from another computer on the network via a ssh session (because you cannot debug the display server -gnome-shell- from within the display server itself), then reproduce the crash and type "t t a bt" from gdb to generate a backtrace for all threads.

In case, a more detailed procedure can be found here: https://wiki.gnome.org/Projects/GnomeShell/Debugging

Comment 5 Daniel Rowe 2017-11-16 11:01:39 UTC
Created attachment 1353423 [details]
backtrace captured with gdb

Thread 1 "gnome-shell" received signal SIGSEGV, Segmentation fault.
meta_logical_monitor_get_scale (logical_monitor=logical_monitor@entry=0x0) at backends/meta-logical-monitor.c:181
181	  return logical_monitor->scale;

Comment 6 Daniel Rowe 2017-11-16 11:10:37 UTC
More info:

This is on a intel NUC with Intel(R) Core(TM) i5-6260U CPU

Monitor is connected with a display port cable and is 4k.

Comment 7 Daniel Rowe 2017-11-16 11:31:08 UTC
Same bug as this?

https://bugzilla.redhat.com/show_bug.cgi?id=1500325

Comment 8 Olivier Fourdan 2017-11-16 12:22:01 UTC
(In reply to Daniel Rowe from comment #7)
> Same bug as this?
> 
> https://bugzilla.redhat.com/show_bug.cgi?id=1500325

Yes, most likely, however to be sure we'd need a complete backtrace (all the steps), not just the step where the crash occurs (comment 5)

This particular issue is reported upstream as bug https://bugzilla.gnome.org/show_bug.cgi?id=788764 and already fixed in git master for mutter.

Comment 9 biltong 2017-11-16 20:37:29 UTC
This is hitting me whenever the screensaver kicks in. Losing all my work each time.

Comment 10 Daniel Rowe 2017-11-17 23:51:13 UTC
So the backtrace attached to comment 5 OK?

Comment 11 Olivier Fourdan 2017-11-20 08:17:21 UTC
(In reply to Daniel Rowe from comment #10)
> So the backtrace attached to comment 5 OK?

It's incomplete, it's just the very last step, a backtrace contains all the steps that led  to the crash, i.e. in gdb you need to type "bt" for "backtrace" to get the entire trace.

However, I'd say, there is a good chance this is a known issue already fixed upstream (comment 8)

Comment 12 Daniel Rowe 2017-11-20 08:23:48 UTC
That's what is in the backtrace.log file.

I attached a file rather than direct cut and paste or am I completely confused?

Comment 13 Olivier Fourdan 2017-11-20 10:09:43 UTC
(In reply to Daniel Rowe from comment #12)
> That's what is in the backtrace.log file.
> 
> I attached a file rather than direct cut and paste or am I completely
> confused?

Oh, I'm completely sorry, I didn't see the attachment! So yes, absolutely, this is perfectly OK.

Comment 14 Olivier Fourdan 2017-11-20 17:10:47 UTC
So yes, your backtrace is similar to the one in https://bugzilla.gnome.org/page.cgi?id=traceparser/trace.html&trace_id=238066 meaning that the fix for  https://bugzilla.gnome.org/show_bug.cgi?id=788764 will fix that issue as well.

Comment 15 biltong 2017-11-21 17:27:10 UTC
Can you see how many other people are experiencing this bug?

Comment 16 biltong 2017-11-22 18:42:13 UTC
Any chance of a backport for this? I will re-install Fedora 26 if I have to. Can't rebuild mutter myself because of bug 1507661

Comment 17 yaro014 2017-12-11 10:34:16 UTC
I have the same issue.First tried latest beta fedora 27 some time ago. Whenever the screen locked itself, any tries to login back were ending up spawning second gnome session after which the previous one crashed. Sometimes it helped to go between terminal session and back to X and it would log you back in to your old running session. This disappeared for me with fedora 27 stable. Issues came back ~weeks ago. 

fpaste --sysinfo : https://paste.fedoraproject.org/paste/HvnZx-ql8qKCZNgvTN4EVA

Same issue on three different configurations

1. Laptop thinnkpad p50 intel+novaueu 
2. Desktop AMD APU
3. Desktop AMD APU + Nvidia dedicated graphics (novaueu)

Comment 18 rugk 2017-12-12 23:16:05 UTC
Hmm… is this issue maybe related (or a duplicate of) https://bugzilla.redhat.com/show_bug.cgi?id=1516466?

Comment 19 sam0hack 2018-01-27 16:10:51 UTC
When i insert the My Pendrive Gnome Creash and take me to the login screen. I test it with 2 pendrives both are NTFS
 Formated and Working Good in another laptop.I tested this several times and everytime it crash only when insert
 usb drive However usb mouse working good.

Comment 20 ALI-S0 2018-01-28 16:27:49 UTC
When I connect my phone to the computer via USB, Gnome crashed and took me to the login screen, I try with another mobile and got the same result

Comment 21 sam0hack 2018-01-28 16:47:52 UTC
(In reply to ALI-S0 from comment #20)
> When I connect my phone to the computer via USB, Gnome crashed and took me
> to the login screen, I try with another mobile and got the same result

Yes, it's been 4 days but not fixed.I tried re-installing GDM and Wayland but the problem not fixed. So I installed Cinnamon Desktop and choose cinnamon (software render) when login And it is working just fine.

Comment 22 ALI-S0 2018-02-01 11:47:09 UTC
(In reply to sam0hack from comment #21)
> (In reply to ALI-S0 from comment #20)
> > When I connect my phone to the computer via USB, Gnome crashed and took me
> > to the login screen, I try with another mobile and got the same result
> 
> Yes, it's been 4 days but not fixed.I tried re-installing GDM and Wayland
> but the problem not fixed. So I installed Cinnamon Desktop and choose
> cinnamon (software render) when login And it is working just fine.

I Stopped `places status indicator` extension and gnome not crashed, i tried to mount and unmount flash usb many time and no crash happened

Comment 23 sam0hack 2018-02-03 08:55:42 UTC
(In reply to wsa90 from comment #22)
> (In reply to sam0hack from comment #21)
> > (In reply to ALI-S0 from comment #20)
> > > When I connect my phone to the computer via USB, Gnome crashed and took me
> > > to the login screen, I try with another mobile and got the same result
> > 
> > Yes, it's been 4 days but not fixed.I tried re-installing GDM and Wayland
> > but the problem not fixed. So I installed Cinnamon Desktop and choose
> > cinnamon (software render) when login And it is working just fine.
> 
> I Stopped `places status indicator` extension and gnome not crashed, i tried
> to mount and unmount flash usb many time and no crash happened

I stopped places status indicator extension and it is not crashing anymore.
Thanks @wsa90

Comment 24 ALI-S0 2018-02-03 11:04:31 UTC
(In reply to sam0hack from comment #23)
> I stopped places status indicator extension and it is not crashing anymore.
> Thanks @wsa90

You're welcome

Comment 25 Ben Cotton 2018-11-27 17:14:20 UTC
This message is a reminder that Fedora 27 is nearing its end of life.
On 2018-Nov-30  Fedora will stop maintaining and issuing updates for
Fedora 27. It is Fedora's policy to close all bug reports from releases
that are no longer maintained. At that time this bug will be closed as
EOL if it remains open with a Fedora  'version' of '27'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora 27 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora, you are encouraged  change the 'version' to a later Fedora 
version prior this bug is closed as described in the policy above.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events. Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

Comment 26 Ben Cotton 2018-11-30 19:35:34 UTC
Fedora 27 changed to end-of-life (EOL) status on 2018-11-30. Fedora 27 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.


Note You need to log in before you can comment on or make changes to this bug.