at-spi-bus-launcher segfaults

girzel · 2022-04-12 23:22:36

I've had trouble with the at-spi-dbus-bus service failing for quite some time, causing many applications to time out while waiting for its dbus bus. This is on a machine with no desktop environment, running X11 and i3.

I poked into it, and tried running

/usr/lib/at-spi-bus-launcher

directly, and it seemed to hang. But when I looked at the journal, it actually appeared to be segfaulting.

The first line is the normal attempt to start the bus, in this case requested by Firefox. The bus fails, meaning that everything that tries to use it needs to wait for the default Dbus timeout of whatever it is, 20 seconds or so. The rest of the log seems to be the result of me trying to run the launcher manually (as my user):

Apr 11 08:46:40 clem dbus-daemon[932]: [session uid=1000 pid=932] Activating systemd to hand-off: service name='org.a11y.Bus' unit='at-spi-dbus-bus.service' requested by ':1.3' (uid=1000 pid=1589 comm="/usr/lib/firefox/firefox")
Apr 12 15:08:24 clem audit[107795]: ANOM_ABEND auid=1000 uid=1000 gid=1000 ses=1 pid=107795 comm="at-spi-bus-laun" exe="/usr/lib/at-spi-bus-launcher" sig=5 res=1
Apr 12 15:08:24 clem kernel: traps: at-spi-bus-laun[107795] trap int3 ip:7fe118d37540 sp:7ffcf23c4200 error:0 in libglib-2.0.so.0.7200.0[7fe118cf8000+92000]
Apr 12 15:08:24 clem kernel: audit: type=1701 audit(1649801304.630:1690): auid=1000 uid=1000 gid=1000 ses=1 pid=107795 comm="at-spi-bus-laun" exe="/usr/lib/at-spi-bus-launcher" sig=5 res=1
Apr 12 15:08:24 clem systemd-coredump[107797]: Process 107795 (at-spi-bus-laun) of user 1000 dumped core.
                                               
                                               Module linux-vdso.so.1 with build-id ae8518a0710c2679844504efb751b01abf13cef5
                                               Module libXdmcp.so.6 with build-id 8ca0792d23c8b8b4c0864297512349292bea5955
                                               Module libXau.so.6 with build-id 1c67764663e07bec24d8951e5fd93f4d165979ff
                                               Module libgpg-error.so.0 with build-id f9aaa96d5ddf6f6872051ed7abcdfe6bdaff9b1c
                                               Module libpthread.so.0 with build-id 7fa8b52fae071a370ba4ca32bf9490a30aff31c4
                                               Module libblkid.so.1 with build-id 140694a62d8d4d07c6c320a501f948dd1b389d73
                                               Module libxcb.so.1 with build-id 0d1ef11740a5daad2ee331e812a51aa6574af222
                                               Module ld-linux-x86-64.so.2 with build-id c09c6f50f6bcec73c64a0b4be77eadb8f7202410
                                               Module libgcc_s.so.1 with build-id 5d817452a709ca3a213341555ddcf446ecee37fa
                                               Module libgcrypt.so.20 with build-id 8bf3cb884124273640de797a3e77d86c98434ea4
                                               Module libcap.so.2 with build-id eb6dae97527fc89dbb0d5bb581a15acd02ae9f56
                                               Module liblz4.so.1 with build-id e63600ab23b2f6997f42fac2fa56e1f02ce159a1
                                               Module libzstd.so.1 with build-id 3bccb8fe08e48d5ea135b1d0f99de0d771dd752f
                                               Module liblzma.so.5 with build-id 28b40c7af8098a66af6ee093b6986b91cad7694d
                                               Module libpcre.so.1 with build-id 845483dd0acba86de9f0313102bebbaf3ce52767
                                               Module libffi.so.8 with build-id f0a9586cf0f42d2b9971bd1065ca3a6b19f4a2c2
                                               Module libmount.so.1 with build-id 4436aeea0cd8c01b5a77969e0531184f8b3513ce
                                               Module libz.so.1 with build-id 1fb800ce60ddb605ebe23f9702adcd341c7c8970
                                               Module libgmodule-2.0.so.0 with build-id 5d0db204364cefb16d6d80f9e40df7c3d86023b3
                                               Module libc.so.6 with build-id 85766e9d8458b16e9c7ce6e07c712c02b8471dbc
                                               Module libX11.so.6 with build-id 1bf9c2f1c9e2f823a1200b41ffa6b09943edf886
                                               Module libsystemd.so.0 with build-id 05d0460efefd339d315812f9506aecc5a5d22200
                                               Module libglib-2.0.so.0 with build-id d6c7c03d71a1b71f59e10016323136de55f43266
                                               Module libgobject-2.0.so.0 with build-id f5126c30685462884948f1048f2039305c67f5c5
                                               Module libgio-2.0.so.0 with build-id 3f16bee59e25c8bfbb70c4e78a3c90ee79ba4469
                                               Module at-spi-bus-launcher with build-id d691951cad1c81fe23ebe81c48d9f2901b58f0f9
                                               Stack trace of thread 107795:
                                               #0  0x00007fe118d37540 g_log_structured_array (libglib-2.0.so.0 + 0x5c540)
                                               #1  0x00007fe118d3780a g_log_default_handler (libglib-2.0.so.0 + 0x5c80a)
                                               #2  0x00007fe118d38cdd g_logv (libglib-2.0.so.0 + 0x5dcdd)
                                               #3  0x00007fe118d38fe4 g_log (libglib-2.0.so.0 + 0x5dfe4)
                                               #4  0x0000560cdfc2e454 n/a (at-spi-bus-launcher + 0x2454)
                                               #5  0x00007fe1188de310 __libc_start_call_main (libc.so.6 + 0x2d310)
                                               #6  0x00007fe1188de3c1 __libc_start_main@@GLIBC_2.34 (libc.so.6 + 0x2d3c1)
                                               #7  0x0000560cdfc2e485 n/a (at-spi-bus-launcher + 0x2485)
                                               ELF object binary architecture: AMD x86-64

seth · 2022-04-13 06:47:59

If you don't need the a11y stuff, you can just

export NO_AT_BRIDGE=1

The backtrace pattern seems linked to incompatible extensions (gnome update…) or gnome/gtk related AUR packages (plugins?) that need to be rebuilt.

girzel · 2022-04-13 15:08:36

Thanks for the reply! I forgot to mention that about halfway through this process I did set NO_AT_BRIDGE=1, in /etc/environment, and it hasn't had any effect. The bus creation still fails, and other applications still wait on it.

I do have a sizable collection of gnome/gtk AUR packages by this point, I'll go through them and see what I can uninstall outright, and try rebuilding the rest. Thanks for the tip.

seth · 2022-04-13 15:28:42

Sure the environment is sourced (correctly)?
I'm exporting it in my .xinitrc (sourcing .xprofile and exporting it there) and there's no at-spi-launcher process running nor attempted anymore ever since…

girzel · 2022-04-14 16:26:27

Hmm. (Maybe unsurprisingly) the problem doesn't seem to be what I originally thought it was. You're right I wasn't setting NO_AT_BRIDGE correctly, but once I got that sorted the timeout started to come from an inability to connect to ibus (which Zoom has as a dependency for some reason), and once I succeeded in uninstalling ibus, the error was a timeout in StartServiceByName for org.gtk.vfs.Daemon. I'm testing this by starting Geeqie from the command line, but it looks like the same situation when I start Libreoffice, or get anything using GTK to open a file-chooser dialog, etc. Here's the Geeqie timeout:

[eric@clem ~]$ geeqie

(geeqie:3533): GVFS-WARNING **: 09:17:06.529: Error creating proxy: Error calling StartServiceByName for org.gtk.vfs.Daemon: Timeout was reached (g-io-error-quark, 24)

So obviously something more fundamental is going wrong with... Dbus? And GTK? I looked for failed systemd units: nothing at the system level, under the user level there's only pulseaudio and redshift failed, neither of which seem relevant. I've got GTK2, 3 and 4 installed (love the dependencies), so I guess it's not surprising that things are fighting each other.

I no longer have any clues! Do you know where I could look next?

seth · 2022-04-14 17:19:00

"loginctl session-status" - and likely see the last link below...

girzel · 2022-04-17 17:24:47

My ~/.xinitrc doesn't have anything in it but setting some environment variables and then "exec /usr/bin/i3".

Here's my session-status, I guess I don't see anything very notable here...

py/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73388 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73389 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73391 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73392 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73439 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73440 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73467 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73468 /usr/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73477 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73478 /usr/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73486 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73487 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73496 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73497 /usr/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73503 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73504 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73507 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73508 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73518 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73519 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73522 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73523 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─73532 /home/eric/.emacs.d/elpy/rpc-venv/bin/python -W ignore -m elpy.__main__
		  ├─73533 /home/eric/.cache/pypoetry/virtualenvs/unzipupload-aVRtSUBE-py3.9/bin/python /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages/jedi/inference/compiled/subprocess/__main__.py /home/eric/.emacs.d/elpy/rpc-venv/lib/python3.10/site-packages 3.10.4
		  ├─80149 /usr/lib/firefox/firefox -contentproc -childID 264 -isForBrowser -prefsLen 9900 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─80259 /usr/lib/firefox/firefox -contentproc -childID 266 -isForBrowser -prefsLen 9901 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─80304 /usr/lib/firefox/firefox -contentproc -childID 267 -isForBrowser -prefsLen 9901 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─80395 /usr/lib/firefox/firefox -contentproc -childID 268 -isForBrowser -prefsLen 9901 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─80452 /usr/lib/firefox/firefox -contentproc -childID 270 -isForBrowser -prefsLen 9901 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─83277 /usr/lib/firefox/firefox -contentproc -childID 277 -isForBrowser -prefsLen 9900 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─83668 /usr/lib/firefox/firefox -contentproc -childID 280 -isForBrowser -prefsLen 9901 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─84458 /usr/lib/firefox/firefox -contentproc -childID 282 -isForBrowser -prefsLen 9900 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─84830 /usr/lib/firefox/firefox -contentproc -childID 283 -isForBrowser -prefsLen 9900 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─84893 /usr/lib/firefox/firefox -contentproc -childID 284 -isForBrowser -prefsLen 9900 -prefMapSize 271447 -jsInitLen 277748 -parentBuildID 20220413030647 -appDir /usr/lib/firefox/browser 1698 true tab
		  ├─85063 pavucontrol
		  └─97250 loginctl session-status

Apr 15 09:00:49 clem sudo[41712]: pam_unix(sudo:session): session opened for user root(uid=0) by (uid=1000)
Apr 15 09:00:49 clem sudo[41712]: pam_unix(sudo:session): session closed for user root
Apr 15 09:00:51 clem sudo[41722]: pam_systemd_home(sudo:account): systemd-homed is not available: Unit dbus-org.freedesktop.home1.service not found.
Apr 15 09:00:51 clem sudo[41722]:     eric : TTY=pts/2 ; PWD=/home/eric/dev/emacs ; USER=root ; COMMAND=/usr/bin/pacman --color=always --sync libgccjit
Apr 15 09:00:51 clem sudo[41722]: pam_unix(sudo:session): session opened for user root(uid=0) by (uid=1000)
Apr 15 09:00:51 clem sudo[41723]: pam_systemd_home(sudo:account): systemd-homed is not available: Unit dbus-org.freedesktop.home1.service not found.
Apr 15 09:00:51 clem sudo[41723]:     eric : TTY=pts/2 ; PWD=/home/eric/dev/emacs ; USER=root ; COMMAND=/usr/bin/pacman -T
Apr 15 09:00:51 clem sudo[41723]: pam_unix(sudo:session): session opened for user root(uid=0) by (uid=1000)
Apr 15 09:00:51 clem sudo[41723]: pam_unix(sudo:session): session closed for user root
Apr 15 09:00:53 clem sudo[41722]: pam_unix(sudo:session): session closed for user root

seth · 2022-04-17 17:30:51

That's not or at least no complete session status and it lacks all the relevant bits.
Don't copy out of the pager, redirect the output into a file and

My ~/.xinitrc doesn't have anything in it but setting some environment variables and then "exec /usr/bin/i3".

is an extremely strong indication that your xinitrc /is/ broken, so post that as well.

girzel · 2022-04-17 18:12:36

seth wrote:

That's not or at least no complete session status and it lacks all the relevant bits.
Don't copy out of the pager, redirect the output into a file and

That's exactly what I did. Are there additional switches I should be passing to "session-status" to get more information?

seth wrote:

an extremely strong indication that your xinitrc /is/ broken, so post that as well.

Here's the xinitrc, including all its glorious historical cruft. I just tried commenting out the fcitx-related variables and rebooting, and it had no effect.

# fcitx runs as a daemon by default, do I need the "&"
export GTK_IM_MODULE=fcitx
export QT_IM_MODULE=fcitx
export XMODIFIERS=@im=fcitx

#echo "UPDATESTARTUPTTY" | gpg-connect-agent > /dev/null 2&>1

#numlockx &

exec /usr/bin/i3

Slithery · 2022-04-17 18:14:26

So it's missing vital sections then.
Please read,,,
https://wiki.archlinux.org/title/Xinit#xinitrc

seth · 2022-04-17 18:17:56

And if that's still not enough provide the session status through

loginctl session-status | curl -F 'file=@-' 0x0.st

girzel · 2022-04-17 18:44:36

Oh! I guess I wasn't paying attention when that recommended config changed. I've now updated my ~/.xinitrc and pasted it at the bottom.

On reboot I still see the same problems, the geeqie test takes 20+ seconds before reaching the timeout below, and then starting.

[eric@clem ~]$ geeqie

(geeqie:1262): GVFS-WARNING **: 11:30:54.698: Error creating proxy: Error calling StartServiceByName for org.gtk.vfs.Daemon: Timeout was reached (g-io-error-quark, 24)

I've piped my session-status up, and yes, it is the same output, just with less in it because I've rebooted recently: http://0x0.st/oc6u.txt

#!/bin/sh

userresources=$HOME/.Xresources
usermodmap=$HOME/.Xmodmap
sysresources=/etc/X11/xinit/.Xresources
sysmodmap=/etc/X11/xinit/.Xmodmap

# merge in defaults and keymaps

if [ -f $sysresources ]; then







    xrdb -merge $sysresources

fi

if [ -f $sysmodmap ]; then
    xmodmap $sysmodmap
fi

if [ -f "$userresources" ]; then







    xrdb -merge "$userresources"

fi

if [ -f "$usermodmap" ]; then
    xmodmap "$usermodmap"
fi

# start some nice programs

if [ -d /etc/X11/xinit/xinitrc.d ] ; then
 for f in /etc/X11/xinit/xinitrc.d/?*.sh ; do
  [ -x "$f" ] && . "$f"
 done
 unset f
fi

exec /usr/bin/i3

# twm &
# xclock -geometry 50x50-1+1 &
# xterm -geometry 80x50+494+51 &
# xterm -geometry 80x20+494-0 &
# exec xterm -geometry 80x66+0+0 -name login

seth · 2022-04-17 19:00:55

And it especially (now) says

	  Leader: 454 (login)
	    Seat: seat0; vc1
	     TTY: tty1
	 Service: login; type tty; class user
	   State: active
	    Unit: session-1.scope
		  ├─ 454 "login -- eric"
		  ├─ 823 -bash
		  ├─1072 /bin/sh /usr/bin/startx

What happens for

/usr/lib/gvfsd --replace

?

Edit: also

paacman -Qikk gvfs

Last edited by seth (2022-04-17 19:03:02)

girzel · 2022-04-17 23:45:38

seth wrote:

What happens for /usr/lib/gvfsd --replace

Was that supposed to be run as my user or root? As my user it hangs indefinitely, as root it gives me:

[eric@clem ~]$ sudo /usr/lib/gvfsd --replace
Failed to connect to the D-BUS daemon: Error spawning command line “dbus-launch --autolaunch=9cadfb2128b346a993b168dbc62e34ea --binary-syntax --close-stderr”: Child process exited with code 1 (g-spawn-exit-error-quark, 1)]

Edit: also
paacman -Qikk gvfs

That gives me:

[eric@clem ~]$ pacman -Qikk gvfs
Name            : gvfs
Version         : 1.50.0+r5+g747c7f6e-1
Description     : Virtual filesystem implementation for GIO
Architecture    : x86_64
URL             : https://wiki.gnome.org/Projects/gvfs
Licenses        : LGPL
Groups          : gnome
Provides        : None
Depends On      : avahi  dconf  fuse3  libarchive  libcdio-paranoia  libsoup3  udisks2  libsecret
                  libbluray  libgudev  gcr  psmisc  gsettings-desktop-schemas
Optional Deps   : gvfs-afc: AFC (mobile devices) support
                  gvfs-smb: SMB/CIFS (Windows client) support
                  gvfs-gphoto2: gphoto2 (PTP camera/MTP media player) support
                  gvfs-mtp: MTP device support
                  gvfs-goa: gnome-online-accounts (e.g. OwnCloud) support
                  gvfs-nfs: NFS support
                  gvfs-google: Google Drive support
                  gtk3: Recent files support [installed]
Required By     : gimp
Optional For    : ffmpegthumbnailer  inkscape  thunar
Conflicts With  : None
Replaces        : gvfs-obexftp  gvfs-afp
Installed Size  : 5.23 MiB
Packager        : Jan Alexander Steffens (heftig) <heftig@archlinux.org>
Build Date      : Wed 13 Apr 2022 09:31:35 AM PDT
Install Date    : Thu 14 Apr 2022 08:49:46 AM PDT
Install Reason  : Installed as a dependency for another package
Install Script  : No
Validated By    : Signature

warning: gvfs: /usr/share/polkit-1/rules.d/org.gtk.vfs.file-operations.rules (Permission denied)
gvfs: 324 total files, 1 altered file

Looks like a clue...

seth · 2022-04-18 06:45:36

No, everthing is fine w/ the package and gvfsd doesn't so much hang as that it simply runs and the sudo call fails expectably because of the missing session bus in that environment.
There's no problem indicated here

After running gvfsd (as user), does geeqie startup timely?
Also

printenv | curl -F 'file=@-' 0x0.st

girzel · 2022-04-18 16:06:09

seth wrote:

After running gvfsd (as user), does geeqie startup timely?

Yes! While gvfsd is running geeqie starts instantly. When I interrupt gvfsd it goes right back to the earlier behavior.

Output of printenv: http://0x0.st/ocGD.txt

Edit: Further random information: I have "gvfs" and "gvfsd" directories under "/run/user/1000", but neither have anything in them (even while the manual gvfsd process is running). Is that significant?

Last edited by girzel (2022-04-18 16:09:24)

seth · 2022-04-18 20:11:31

Nope, that's normal. The DBUS_SESSION_BUS_ADDRESS is fine, too…
errno 24 is EMFILE "Too many open files" when trying to dbus-activate gvfsd.

This is just a hunc from

XDG_DATA_DIRS=/home/eric/.local/share/flatpak/exports/share:/var/lib/flatpak/exports/share:/usr/local/share/:/usr/share/

, but

pacman -Qs portal

girzel · 2022-04-18 21:01:00

seth wrote:

pacman -Qs portal

That gives me:

local/xdg-desktop-portal 1.14.3-1
    Desktop integration portals for sandboxed apps
local/xdg-desktop-portal-gtk 1.14.0-1
    A backend implementation for xdg-desktop-portal using GTK

I don't even know when/why I installed flatpak, this is my "install all the packages" machine. If I look into uninstalling it:

[eric@clem ~]$ pikaur -Rs flatpak flatpak-builder
checking dependencies...
:: firefox optionally requires xdg-desktop-portal: Screensharing with Wayland
:: myrepos optionally requires bzr: support for bzr repositories
:: pipewire optionally requires rtkit: realtime privileges with rtkit module

Packages (11) appstream-glib-0.7.18-2  breezy-3.2.1-2  gcab-1.4-1  ostree-2022.2-1
              python-configobj-5.0.6.r110.g3e2f4cc-3  python-patiencediff-0.2.2-3  rtkit-0.13-1
              xdg-desktop-portal-1.14.3-1  xdg-desktop-portal-gtk-1.14.0-1  flatpak-1:1.12.7-1
              flatpak-builder-1.2.2-1

Total Removed Size:  58.80 MiB

:: Do you want to remove these packages? [Y/n]

I don't see anything in there I care very much about. Should I just pull all that out?

Thanks for sticking with this...

girzel · 2022-04-18 23:00:47

None of that looked very important, so I just uninstalled and rebooted. Same behavior, unfortunately.

My sense is that this isn't really a gvfs problem. The way the error reporting moved from at-spi, to ibus, to gvfs makes me think that the real issue is underlying all those (something to do with the user-level dbus bus?), and by fixing/removing those other packages, I'm just going down the line to the next thing to trigger the error.

seth · 2022-04-19 06:15:57

The problem is the dbus activation of that service, and the xdg-portal* services have historically shown to cause trouble around this.

pkill gvfs
dbus-monitor | tee /tmp/dbus.debug

And in a second terminal

dbus-send --session --print-reply --dest=org.gtk.vfs.Daemon /org/freedesktop/DBus org.freedesktop.DBus.Introspectable.Introspect

This should™ activate the service by running gvfsd or maybe yell some error at you. In addition we monitor the bus in /tmp/dbus.debug

girzel · 2022-04-19 15:29:04

seth wrote:

The problem is the dbus activation of that service, and the xdg-portal* services have historically shown to cause trouble around this.

Okay, I no longer have any packages with the name "portal" installed on my system.

seth wrote:

pkill gvfs
dbus-monitor | tee /tmp/dbus.debug
And in a second terminal
dbus-send --session --print-reply --dest=org.gtk.vfs.Daemon /org/freedesktop/DBus org.freedesktop.DBus.Introspectable.Introspect
This should™ activate the service by running gvfsd or maybe yell some error at you. In addition we monitor the bus in /tmp/dbus.debug

Here's the result in /tmp/dbus.debug: http://0x0.st/ochV.txt

After I ran the code above I launched Libreoffice, which displays the same behavior as Geeqie: it sits on the splash screen for 20 seconds or so before booting. In the terminal window with "dbus-send", I got:

[eric@clem ~]$ dbus-send --session --print-reply --dest=org.gtk.vfs.Daemon /org/freedesktop/DBus org.freedesktop.DBus.Introspectable.Introspect
Error org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.

seth · 2022-04-19 17:33:59

Did you forget or fail to pkill gvfs?
There's no "string "Could not get owner of name 'org.gtk.vfs.Daemon': no such name"" that would lead to the daemon being activated.
Maybe there's a stale process?

pkill gvfs; ps aux | grep gvfs

cinnaboot · 2022-04-19 19:26:18

Hi, I'm also following along, but have different output from the dbus-send command: http://ix.io/3Vyz

I'm not sure if I have the same problem, or should make a new topic though: at-spi-bus-launcher is preventing systemd from restarting unless I manually kill it. Have tried the previous suggestions in the thread.

girzel · 2022-04-19 20:44:28

seth wrote:

Did you forget or fail to pkill gvfs?
There's no "string "Could not get owner of name 'org.gtk.vfs.Daemon': no such name"" that would lead to the daemon being activated.
Maybe there's a stale process?
pkill gvfs; ps aux | grep gvfs

There's never a gvfs process running unless I do the explicit "/usr/lib/gvfsd --replace" thing. I did run the pkill command, and confirmed that there's no gvfs process running. Maybe it's time to reboot and try everything from scratch...

seth · 2022-04-21 14:05:21

Can you manually start the service?

systemctl --user start gvfs-daemon
systemctl --user status gvfs-daemon

If yes, as a workaround you could simply enable the service,

systemctl --user enable gvfs-daemon

Arch Linux

#1 2022-04-12 23:22:36

at-spi-bus-launcher segfaults

#2 2022-04-13 06:47:59

Re: at-spi-bus-launcher segfaults

#3 2022-04-13 15:08:36

Re: at-spi-bus-launcher segfaults

#4 2022-04-13 15:28:42

Re: at-spi-bus-launcher segfaults

#5 2022-04-14 16:26:27

Re: at-spi-bus-launcher segfaults

#6 2022-04-14 17:19:00

Re: at-spi-bus-launcher segfaults

#7 2022-04-17 17:24:47

Re: at-spi-bus-launcher segfaults

#8 2022-04-17 17:30:51

Re: at-spi-bus-launcher segfaults

#9 2022-04-17 18:12:36

Re: at-spi-bus-launcher segfaults

#10 2022-04-17 18:14:26

Re: at-spi-bus-launcher segfaults

#11 2022-04-17 18:17:56

Re: at-spi-bus-launcher segfaults

#12 2022-04-17 18:44:36

Re: at-spi-bus-launcher segfaults

#13 2022-04-17 19:00:55

Re: at-spi-bus-launcher segfaults

#14 2022-04-17 23:45:38

Re: at-spi-bus-launcher segfaults

#15 2022-04-18 06:45:36

Re: at-spi-bus-launcher segfaults

#16 2022-04-18 16:06:09

Re: at-spi-bus-launcher segfaults

#17 2022-04-18 20:11:31

Re: at-spi-bus-launcher segfaults

#18 2022-04-18 21:01:00

Re: at-spi-bus-launcher segfaults

#19 2022-04-18 23:00:47

Re: at-spi-bus-launcher segfaults

#20 2022-04-19 06:15:57

Re: at-spi-bus-launcher segfaults

#21 2022-04-19 15:29:04

Re: at-spi-bus-launcher segfaults

#22 2022-04-19 17:33:59

Re: at-spi-bus-launcher segfaults

#23 2022-04-19 19:26:18

Re: at-spi-bus-launcher segfaults

#24 2022-04-19 20:44:28

Re: at-spi-bus-launcher segfaults

#25 2022-04-21 14:05:21

Re: at-spi-bus-launcher segfaults

Board footer