GLib: g_variant_new_string: assertion 'string != NULL' failed

Bug #1740522 reported by Federated Jam
42
This bug affects 9 people
Affects Status Importance Assigned to Milestone
gdm
Expired
Medium
gdm3 (Ubuntu)
Confirmed
Low
Unassigned

Bug Description

I saw this in the Logs application

gdm3
GLib: g_variant_new_string: assertion 'string != NULL' failed

ProblemType: Bug
DistroRelease: Ubuntu 17.10
Package: gdm3 3.26.1-3ubuntu3
Uname: Linux 4.14.7-041407-generic x86_64
ApportVersion: 2.20.7-0ubuntu3.6
Architecture: amd64
CurrentDesktop: ubuntu:GNOME
Date: Fri Dec 29 17:01:28 2017
InstallationDate: Installed on 2017-12-19 (9 days ago)
InstallationMedia: Ubuntu 17.10 "Artful Aardvark" - Release amd64 (20171018)
ProcEnviron:
 LANGUAGE=en_GB:en
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=en_GB.UTF-8
 SHELL=/bin/bash
SourcePackage: gdm3
UpgradeStatus: No upgrade log present (probably fresh install)
mtime.conffile..etc.gdm3.custom.conf: 2017-12-19T18:13:29.216565

Revision history for this message
Federated Jam (federated-jam) wrote :
Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Sounds like a problem with gnome-terminal:

Dec 27 12:51:56 hostname gnome-terminal-[5255]: g_variant_new_string: assertion 'string != NULL' failed
Dec 27 12:51:56 hostname gnome-terminal-[5255]: g_variant_new_variant: assertion 'value != NULL' failed
Dec 27 12:51:56 hostname gnome-terminal-[5255]: g_variant_get_type: assertion 'value != NULL' failed

What makes you say it's a gdm3 bug?

Changed in gdm3 (Ubuntu):
status: New → Incomplete
Revision history for this message
Federated Jam (federated-jam) wrote :

According to the Logs application, the sender of the message was gdm3, so it seemed a reasonable guess.

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Probably both are true... A quick google shows many apps experiencing that same assertion failure.

If you'd like to keep this bug about the gdm3 case then we'll need some log file(s) (or screenshots?) of the problem being reported against gdm3. I'm hoping for a little bit more contextual information.

Revision history for this message
Federated Jam (federated-jam) wrote :

I also saw many comments about this via the web, but didn't see any bugs people could track the situation with.

The screenshot that informed my report is attached.

If there's any more information that will help find the root cause, I'm happy to provide it.

Revision history for this message
Federated Jam (federated-jam) wrote :

It would be an easy fix if the assertion message printed the backtrace of the apps calling in glib.

I wonder if the system packages are compiled in RelWithDbgInfo by default, so that a stacktrace would be useful in release for fixing bugs like this. I wonder if there's a way to extend G_DEBUG with a "slient-stacktraces" option that would report a core file to the system on each critical/warning, because I assume it's in Ubunutu's best interests to fix all of these they can, but of course you don't want all the apps crashing, so G_DEBUG=fatal-criticals doesn't seem it will work.

(Very much a newbie to Ubuntu and GNOME, just wondering...)

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Thanks. Confirmed, but it doesn't seem to be causing any serious problem right now (?)

I can see the error against gdm3 in Ubuntu 18.04 too, by running:

  journalctl /usr/sbin/gdm3

Changed in gdm3 (Ubuntu):
importance: Undecided → Low
status: Incomplete → Confirmed
Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Next step: Please report the issue to the GDM developers using this link:

https://bugzilla.gnome.org/enter_bug.cgi?product=gdm

And when done, tell us the ID of the newly-created bug.

Revision history for this message
Federated Jam (federated-jam) wrote :

> but it doesn't seem to be causing any serious problem right now (?)

At the moment I have an issue where *sometimes* when I return to my computer and the screen has gone to sleep, it takes an incredibly long time to show me the login prompt.

Could be my graphics driver (polaris) or the login manager (gdm) or maybe something I have no idea about it. So maybe it is serious, maybe not.

ID of new bug https://bugzilla.gnome.org/show_bug.cgi?id=792236

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

That is highly likely unrelated to the log messages we're talking about. Please ensure your two issues have two different bug reports.

Revision history for this message
crazyzubr (crazyzubr) wrote :

I have this bug on Ubuntu 18.04. My log for gdm3:

Mar 19 09:50:59 hostname systemd[1]: Stopping GNOME Display Manager...
Mar 19 09:50:59 hostname gdm3[1081]: Freeing conversation 'gdm-password' with active job
Mar 19 09:50:59 hostname gdm3[1081]: Child process -1118 was already dead.
Mar 19 09:50:59 hostname gdm3[1081]: Child process -1118 was already dead.
Mar 19 09:50:59 hostname gdm3[1081]: GLib: g_variant_new_string: assertion 'string != NULL' failed
Mar 19 09:50:59 hostname gdm3[1081]: Failed to contact accountsservice: Ошибка вызова StartServiceByName для org.freedesktop.Accounts: GDBus.Error:org.freedesktop.systemd1.ShuttingDown: Refusing activation,
Mar 19 09:50:59 hostname gdm3[1081]: GLib: g_hash_table_find: assertion 'version == hash_table->version' failed
Mar 19 09:50:59 hostname systemd[1]: Stopped GNOME Display Manager.

Revision history for this message
Oleksandr Vakhovskyi (oleksandr-vakhovskyi) wrote :

I have the same problem after upgrading ubuntu from 17.10 to 18.04. And GUI doesn't start.
What should I do?

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

If this bug is a problem for you then please ask the gdm developers to prioritize it here:
https://bugzilla.gnome.org/show_bug.cgi?id=792236

Changed in gdm:
importance: Unknown → Medium
status: Unknown → Expired
tags: added: bionic
removed: artful
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.