Nixpkgs: Slack crashes intermittently

[utdemir]

Describe the bug
Slack crashes intermittently (once every few hours). I do knot know what triggers it, although it usually happens right after I send a message.

Additional context
Once I ran Slack from the command line and here's the log at the time of the crash:

[09/19/20, 10:45:05:771] info: [RTM] (T029T9GUA) Processed 1 message event(s) in 16.8 ms 
[09/19/20, 10:45:05:775] info: [COUNTS] (T029T9GUA) Updated unread_cnt for [REDACTED]: 1 
[09/19/20, 10:45:05:783] info: [UNREAD-COUNTS] (T029T9GUA) Fetching history for unread counting: D01674UKZ8A, waiting 3927 ms 

(slack:12216): libnotify-WARNING **: 10:45:05.966: Failed to connect to proxy

(slack:12216): libnotify-WARNING **: 10:45:05.968: Failed to connect to proxy

(slack:12216): libnotify-WARNING **: 10:45:05.972: Failed to connect to proxy
[09/19/20, 10:45:05:983] info: Store: NEW_NOTIFICATION {
  [REDACTED]
} 
[09/19/20, 10:45:06:422] info: [DND] (T029T9GUA) There are no members in the presence subscription that we need to fetch DND times for! 
[09/19/20, 10:45:06:504] info: [DESKTOP-MIDDLEWARE] (T029T9GUA) Forwarding redux payload to desktop ["unreads"] 
Cannot upload crash dump: cannot exec /usr/bin/wget

Unexpected crash report id length
Failed to get crash dump id.
Report Id: 
zsh: segmentation fault (core dumped)  slack -f
slack -f  34.89s user 7.46s system 0% cpu 14:32:00.44 total

And here's the snippet from journal, seems to be related to libgtk-3:

Sep 19 10:45:06 serenity kernel: Code: 1f 84 00 00 00 00 00 0f 1f 00 48 8b 47 18 48 8b 40 10 c3 0f 1f 80 00 00 00 00 48 8b 47 18 48 8b 78 20 e9 23 2c ea ff 0f 1f 00 <48> 8b >
Sep 19 10:45:06 serenity kernel: slack[12216]: segfault at 18 ip 00007fc9e72b8460 sp 00007ffcf8da6cc8 error 4 in libgtk-3.so.0.2404.17[7fc9e70b4000+35b000]

Notify maintainers
@mmahut

Metadata

 - system: `"x86_64-linux"`
 - host os: `Linux 5.8.7, NixOS, 21.03.20200911.6d4b933 (Okapi)`
 - multi-user?: `yes`
 - sandbox: `no`
 - version: `nix-env (Nix) 3.0pre20200829_f156513`
 - channels(utdemir): `""`
 - nixpkgs: `/nix/store/r54643sms1ziygm6lkxr062pxy2sjggr-yddp3dcz27j1hhnkw40763aplxxmdydz-source`

Using nixpkgs revision: 6d4b93323e7f78121f8d6db6c59f3889aa1dd931

Maintainer information:

# a list of nixpkgs attributes affected by the problem
attribute: 
  - slack
# a list of nixos modules affected by the problem
module:

[prusnak]

Does the change (version update) in PR https://github.com/NixOS/nixpkgs/pull/98384 fix the issue for you?

[utdemir]

@prusnak , I tried it, but unfortunately got the same error after a few hours:

Sep 22 00:33:19 serenity kernel: Code: 1f 84 00 00 00 00 00 0f 1f 00 48 8b 47 18 48 8b>
Sep 22 00:33:19 serenity kernel: slack[31795]: segfault at 18 ip 00007f96298a4460 sp 0>

and

...
[09/21/20, 23:23:24:169] info: [DND] (T029T9GUA) Will check for changes in DND status again in 5 minutes 
Cannot upload crash dump: cannot exec /usr/bin/wget

Unexpected crash report id length
Failed to get crash dump id.
Report Id: 
zsh: segmentation fault (core dumped)  /nix/store/7ydgdz705wrnfqv9s2rvb7k18kjcjca3-slack-4.9.1/bin/slack
/nix/store/7ydgdz705wrnfqv9s2rvb7k18kjcjca3-slack-4.9.1/bin/slack  64.24s user 9.94s system 0% cpu 14:13:16.33 total

[gsnewmark]

This error kinda reproduces for me after the update to 20.09, but it never crashed before in 20.03:

[пн лис  2 18:58:09 2020] slack[15615]: segfault at 18 ip 00007ff5d44ba460 sp 00007ffcd3913738 error 4 in libgtk-3.so.0.2404.17[7ff5d42b6000+35b000]
[пн лис  2 18:58:09 2020] Code: 1f 84 00 00 00 00 00 0f 1f 00 48 8b 47 18 48 8b 40 10 c3 0f 1f 80 00 00 00 00 48 8b 47 18 48 8b 78 20 e9 23 2c ea ff 0f 1f 00 <48> 8b 47 18 48 8b 40 20 c3 0f 1f 80 00 00 00 00 48 83 ec 08 e8 27

What I've noticed is that the crash happens right after I receive a message (not all of them though, today it happened only twice). I can see the notification for message, but right after it the Slack crashes.

[quidome]

I experience similar crashes in both element and rambox. These are electron apps and they both crash on libgtk:

segfault at 18 ip 00007f8256af5460 sp 00007fffdef349e8 error 4 in libgtk-3.so.0.2404.17[7f82568f1000+35b000]
segfault at 18 ip 00007f5ce0a68460 sp 00007ffc3d391cf8 error 4 in libgtk-3.so.0.2404.17[7f5ce0864000+35b000]

[gsnewmark]

I experience similar crashes in both element and rambox. These are electron apps and they both crash on libgtk:

After my previous comment I've also noticed that other electron apps crash too, for instance skypeforlinux:

лис 16 10:42:22 hyperion kernel: skypeforlinux[30548]: segfault at 18 ip 00007fad30350460 sp 00007ffd7aff3838 error 4 in libgtk-3.so.0.2404.17[7fad3014c000+35b000]

and Discord

лис 11 22:13:31 hyperion kernel: .Discord-wrappe[6454]: segfault at 18 ip 00007f7c6892d460 sp 00007ffe75e82f58 error 4 in libgtk-3.so.0.2404.17[7f7c68729000+35b000]

[djmaze]

Same here since I upgraded to 20.09 yesterday (.electron-wrapp is element-desktop).

$ dmesg | grep segfault
[28494.389639] .electron-wrapp[18396]: segfault at 18 ip 00007f01ee5d2460 sp 00007ffdf2851958 error 4 in libgtk-3.so.0.2404.17[7f01ee3ce000+35b000]
[29995.025882] .electron-wrapp[19914]: segfault at 18 ip 00007fd9110da460 sp 00007fff4997a748 error 4 in libgtk-3.so.0.2404.17[7fd910ed6000+35b000]
[44051.881705] .electron-wrapp[12143]: segfault at 18 ip 00007faf031dc460 sp 00007ffd82d8e6c8 error 4 in libgtk-3.so.0.2404.17[7faf02fd8000+35b000]
[44645.434741] .electron-wrapp[23153]: segfault at 18 ip 00007f5ff3c71460 sp 00007ffc9222bfa8 error 4 in libgtk-3.so.0.2404.17[7f5ff3a6d000+35b000]
[54613.264372] .electron-wrapp[1170]: segfault at 18 ip 00007fda050ca460 sp 00007fffd7405678 error 4 in libgtk-3.so.0.2404.17[7fda04ec6000+35b000]
[61941.284581] .electron-wrapp[14655]: segfault at 18 ip 00007f53b1bcb460 sp 00007ffc1528fd38 error 4 in libgtk-3.so.0.2404.17[7f53b19c7000+35b000]
[68094.235122] teams[7790]: segfault at 18 ip 00007f3af3439460 sp 00007ffcde6676f8 error 4 in libgtk-3.so.0.2404.17[7f3af3235000+35b000]
[68615.769614] mattermost-desk[7357]: segfault at 18 ip 00007fa1fe1a6460 sp 00007ffcc556c178 error 4 in libgtk-3.so.0.2404.17[7fa1fdfa2000+35b000]

[djmaze]

As there is a temporal relation, I am adding that volumeicon and nm-applet began crashing as well, but in libglib:

[67558.407595] traps: volumeicon[28721] trap int3 ip:7f010f6893f5 sp:7ffc62410940 error:0 in libglib-2.0.so.0.6400.5[7f010f64f000+81000]
[67558.424783] traps: .nm-applet-wrap[11177] trap int3 ip:7fe3787293f5 sp:7ffea5eb4930 error:0 in libglib-2.0.so.0.6400.5[7fe3786ef000+81000]

[snicket2100]

I can confirm the same problem on 20.09. In my case it also affects element-desktop, as well as nm-applet, pasystray and syncthing-tray. Strong temporal correlation as well.

Nov 20 17:38:49 somehost kernel: traps: .nm-applet-wrap[6674] trap int3 ip:7f5a5017f3f5 sp:7ffd5bba5e60 error:0 in libglib-2.0.so.0.6400.5[7f5a50145000+81000]
Nov 20 17:38:49 somehost kernel: traps: .pasystray-wrap[8649] trap int3 ip:7f32550473f5 sp:7ffe7f7428c0 error:0 in libglib-2.0.so.0.6400.5[7f325500d000+81000]

Nov 20 17:38:49 somehost .pasystray-wrap[8649]: GdkWindow 0x4400006 unexpectedly destroyed
Nov 20 17:38:49 somehost .pasystray-wrap[8649]: The program 'pasystray' received an X Window System error.
                                               This probably reflects a bug in the program.
                                               The error was 'BadWindow (invalid Window parameter)'.
                                                 (Details: serial 2420 error_code 3 request_code 61 (core protocol) 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 GDK_SYNCHRONIZE environment
                                                  variable to change this behavior. You can then get a meaningful
                                                  backtrace from your debugger if you break on the gdk_x_error() function.)

Nov 20 17:38:49 somehost .nm-applet-wrap[6674]: GdkWindow 0x4000002 unexpectedly destroyed
Nov 20 17:38:49 somehost .nm-applet-wrap[6674]: The program 'nm-applet' received an X Window System error.
                                               This probably reflects a bug in the program.
                                               The error was 'BadWindow (invalid Window parameter)'.
                                                 (Details: serial 12226 error_code 3 request_code 61 (core protocol) 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 GDK_SYNCHRONIZE environment
                                                  variable to change this behavior. You can then get a meaningful
                                                  backtrace from your debugger if you break on the gdk_x_error() function.)

[djmaze]

Are you guys by any chance running i3 as a window manager? I am.

I just ran volumeicon and nm-applet in a terminal in order to see the complete error messages. What I am getting is the RenderBadPicture error as reported in #99197. Which might be fixed with an update of i3. I will try that next.

[gsnewmark]

I'm also getting these errors in i3. I've tried using Gnome yesterday & today and so far errors are gone, and my friend doesn't get any in Plasma. So I'm pretty sure it's caused by either i3 itself or picom (which I use as a compositor).

[quidome]

Same here. I was running i3 when I encountered these issues. I switched to gnome and now they seem to crash no more.

[djmaze]

I just installed i3 from unstable (4.18.3), restarted it and the problem seems gone.