F42 beta not shutting down despite "No inhibitors"

Ask Fedora
, , , ,
1

I am running f42 beta in a VM. Machine is not shutting down when triggering shutdown in Gnome.
Is anyone else experiencing the issue?

Mar 26 11:41:59 vbox-f42 systemd[2231]: Started dbus-:1.2-org.gnome.Ptyxis@1.service.
Mar 26 11:42:01 vbox-f42 systemd[2231]: Started ptyxis-spawn-b66b48cf-0230-407c-896b-50d770e07a42.scope >
Mar 26 11:45:20 vbox-f42 gnome-session[2342]: gnome-session-binary[2342]: WARNING: Client '/org/gnome/Se>
Mar 26 11:45:20 vbox-f42 gnome-session-binary[2342]: WARNING: Client '/org/gnome/SessionManager/Client11>
Mar 26 11:45:20 vbox-f42 gnome-shell[2408]: endSessionDialog: No XDG_SESSION_ID, fetched from logind: 2
Mar 26 11:45:23 vbox-f42 gnome-session-binary[2342]: Entering running state
Mar 26 11:45:23 vbox-f42 gnome-shell[2408]: Gio.IOErrorEnum: GDBus.Error:org.gtk.GDBus.UnmappedGError.Qu>
                                            
                                            Stack trace:
                                              asyncCallback@resource:///org/gnome/gjs/modules/core/overr>
                                              @resource:///org/gnome/shell/ui/init.js:21:20

``
3

Unfortunately the useful stack trace is to the right of what you quoted from the journal. Can you try again to get the full stack trace? You can use --no-pager to avoid opening it in less if that would help. Maybe something like journalctl -e --no-pager | tail. There’s probably a better way…

4

Communicating with Fedora Legal is impractical for Fedora contributors, so what I would do is just ask Fedora Council and expect them to do so if required. If Council says this documentation would be permitted, then let’s do it.

6

@catanzaro is it possible that this post was intended to go elsewhere?

7

Different vm, same result.
journalctl -e --no-pager | tail shows:


Mar 26 21:29:59 vbox-f42 gnome-shell[2226]: endSessionDialog: No XDG_SESSION_ID, fetched from logind: 2
Mar 26 21:30:04 vbox-f42 gnome-session-binary[2148]: Entering running state
Mar 26 21:30:04 vbox-f42 gnome-shell[2226]: Gio.IOErrorEnum: GDBus.Error:org.gtk.GDBus.UnmappedGError.Quark._g_2dio_2derror_2dquark.Code19: Vorgang wurde abgebrochen
                                            
                                            Stack trace:
                                              asyncCallback@resource:///org/gnome/gjs/modules/core/overrides/Gio.js:114:23
                                              @resource:///org/gnome/shell/ui/init.js:21:20

I am not sure how create a better stack trace.

8

Check the output when the issue happens:

gnome-session-inhibit -l

I can reproduce the error, but only with active inhibitors.
Otherwise, reboot and shutdown work correctly for me.

9

It was an unsaved document in the text editor…

However, on the other vm, which currently doesn’t boot (different topic), I am sure there was nothing else open. I report back if I get the “first” vm to boot.

10

Active inhibitors may not be limited to foreground apps, and your problem may be related to VirtualBox.

Also keep in mind that the Beta is already behind the latest updates, so be sure to fully upgrade before trying to reproduce the issue.

11

So, I was able to boot the first vm up again (it wasn’t booting the latest kernel, but the latest-1 one).

On that machine, I can’t reproduce the bug. It must have been an invisible inhibitor, when I hit the bug.

I am wondering if it could be related to the “Install updates” tick box that is presented with the shutdown dialog. I will keep monitoring…

possible too. I fire up a f42-vm in QEMU and see what it does.

12

yes, sigh. Thanks

13

This is a bug in gnome-shell. It’s unfortunate that the stack trace isn’t very good, but you do have the full stack trace, so that’s enough for a bug report. The issue tracker is here; please report it if you can’t find anything similar reported previously. Thank you!

14

I’ve noticed that very occasionally that my laptop with Fedora 42 Beta will look like it shutdown but when I take it out of the backpack the next day the battery is completely empty. The display turns off and it looks “off” but I suspect some small piece of hardware is still using power. The only fix I have it to watch the shutdown text scroll and make sure I see the actual “hardware will power down now” or whatever that final message is. I think if I actually see that then it does shutdown entirely.
The stronger fix I have is to shutdown, then repower up, press Esc to get into the Grub menu and then power off again. That proves that the OS is not involved at all and that I’m powering off via the hardware only (using the physical off switch).

15

done. Gnome 48-rc in F42 unable to shutdown from DE despite "No inhibitors" (#8296) · Issues · GNOME / gnome-shell · GitLab

16

Yes, both machines are updated daily.
I was able to reproduce the bug on my “first” vm.
Machine is not shutting down, gnome-session-inhibit -l says:

No inhibitors

BTW, I think what you experienced with existing inhibitor is also a bug (maybe the [same) - my understanding was that you can still hit the “Shutdown” button and override the inhibitor (for example unsaved text files).

1 Like
18

See also:
Poweroff/reboot doesn't work in GNOME, when an inhibitor is held

If the issue persists, try to stop these services:

rpm -q -l virtualbox-guest-additions | grep -e [.]service$