!XQQVyIbcAcHFvzmcTl:nixos.org

NixOS GNOME

374 Members
A room for maintainers of GNOME & GNOME-adjacent desktop environments (Xfce, Cinnamon, Pantheon...)86 Servers

Load older messages


SenderMessageTime
15 Nov 2025
@capypara:matrix.orgCapyparaYeah the most recent one20:57:14
@capypara:matrix.orgCapyparaAs I said, GNOME doesn't crash, the process just exits with exit code 020:57:24
@capypara:matrix.orgCapyparaah no sorry20:58:45
@capypara:matrix.orgCapypara
Nov 15 21:53:27 kirschbaum systemd[2994]: org.gnome.Shell@wayland.service: Main process exited, code=killed, status=9/KILL
20:58:53
@capypara:matrix.orgCapyparaWas reading the wrong entry, it's SIGKILL20:59:04
@k900:0upti.meK900That sounds like OOM20:59:21
@capypara:matrix.orgCapyparathe kernel has no OOM report in its logs20:59:31
@capypara:matrix.orgCapyparathere is no log about anything unusual in general20:59:38
@k900:0upti.meK900systemd-oomd20:59:39
@capypara:matrix.orgCapypara(in the kernel log)20:59:42
@capypara:matrix.orgCapyparanope21:02:14
@capypara:matrix.orgCapyparaat least it has nothing in it's journal about it killing anything21:02:23
@capypara:matrix.orgCapyparaIt's a mutter bug21:04:51
@capypara:matrix.orgCapyparagot confirmation from the GNOME matrix channel21:04:57
@capypara:matrix.orgCapyparafixed in 49.1.121:04:59
@capypara:matrix.orgCapyparamutter kills it's own process instead of the background app, lol21:05:15
@capypara:matrix.orgCapyparaRedacted or Malformed Event21:07:03
@capypara:matrix.orgCapyparawait, we are on 49.1.121:08:24
@capypara:matrix.orgCapyparafor fucks sake21:18:25
@capypara:matrix.orgCapyparait's a Flatpak bug actually21:18:29
@capypara:matrix.orgCapyparahttps://github.com/flatpak/flatpak/pull/637521:18:30
@capypara:matrix.orgCapyparaCan also reproduce it under Fedora21:19:04
@capypara:matrix.orgCapyparaI just always run into the most cursed bugs somehow.21:19:49
@capypara:matrix.orgCapyparaI don't understand how the Linux desktop seemingly gets more broken every year, like what even21:21:34
@capypara:matrix.orgCapyparaI don't know much about how systemd slices/targets etc. actually work, but I guess the shell is running as PID 0 in some kind of PID namespace? Does systemd have no features to prevent a child process to send a SIGKILL in this context?21:25:56
@k900:0upti.meK900No21:26:22
@k900:0upti.meK900 kill(0) is shortcut for "kill everything" 21:26:36
@capypara:matrix.orgCapyparaAhh right, right, I forgot. That's whats going on, of course.21:26:52
@capypara:matrix.orgCapypara0 kills everything. root PID is 1.21:27:04
@capypara:matrix.orgCapyparaStill crazy that's possible, but yeah I understand why.21:27:22

There are no newer messages yet.


Back to Room ListRoom Version: 6