Bug#981555: does not appear to respond to DBUS interface anymore

Martin Steigerwald Martin.Steigerwald at proact.de
Mon Feb 1 16:12:12 GMT 2021


Resending: I accidentally sent this to Simon directly. Simon nonetheless
kindly replied to me already. I hope I am able to get back to it
tomorrow.


Dear Simon,

Thank you for your detailed reply.

Am Montag, dem 01.02.2021 um 12:47 +0000 schrieb Simon McVittie:
> On Mon, 01 Feb 2021 at 12:06:04 +0100, Martin Steigerwald wrote:
[… Evolution not asking for keyring password anymore …]
> > Since this worked before I do not consider this to be an upstream
> > issue.
> > However also 'gnome-keyring' package does not appear to have been
> > changed
> > recently. So it may be related to the upgrade of a different
> > package.
> 
> I think this might be triggered by security hardening in GLib that
> results
> in gnome-keyring-daemon no longer trusting its parent process to
> supply it
> with a DBUS_SESSION_BUS_ADDRESS. See <https://bugs.debian.org/981420>
> and
> upstream issue <https://gitlab.gnome.org/GNOME/glib/-/issues/2305>.

I see.

> > I am using Runit as PID 1.
> 
> Presumably that means you're getting your D-Bus session bus from dbus-
> x11 (dbus-launch), because the recommended implementation
> dbus-user-session requires systemd?

I have dbus-x11 installed, but dbus-user-session is not installed.

> The various consequences of a non-systemd init, not all of which are
> immediately obvious, are not really something that GNOME projects are
> designed to support.

In addition to other reasons that is one reason I use KDE's Plasma. They
kept Systemd support stuff optional so far.

> The reporter of #981420 *is* using systemd, but one theory about why
> their gnome-keyring isn't working is that they might be using dbus-x11
> rather than dbus-user-session. I haven't been able to confirm or
> disprove that theory yet.

Hmmm, shall I try downgrading to glib 2.66.4-1 from snapshots?

> We try to keep this working in Debian, but there's
> a limit to what we can do in that direction: if the underlying 
> system does not meet various design assumptions then sometimes we will
> just have to say "sorry, you don't have the requirements for this".

Well of course I'd like this to continue to work and if it might be
triggered by security hardening in Glib as you pointed out above, it may
not be related to Systemd at all? A "design assumption" and hard
"requirements" are two different things for me. I prefer software that
works without relying on assumptions.

Thanks,


Mit freundlichen Grüßen / With kind regards
Martin Steigerwald • 
Proact Deutschland GmbH
Trainer
Telefon: +49 911 30999 0 • 
Fax: +49 911 30999 99
Südwestpark 43 • 
90449 Nürnberg • 
Germany
Martin.Steigerwald at proact.de • 
www.proact.de
Amtsgericht Nürnberg
 • 
HRB 18320
Geschäftsführer: 
René Schülein
 • 
Jonas Hasselberg
 • 
Linda Höljö
#ThePowerOfTogether
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/pkg-gnome-maintainers/attachments/20210201/30094e28/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image519114.png
Type: image/png
Size: 60606 bytes
Desc: image519114.png
URL: <http://alioth-lists.debian.net/pipermail/pkg-gnome-maintainers/attachments/20210201/30094e28/attachment-0001.png>


More information about the pkg-gnome-maintainers mailing list