Package: gnome-shell
Version: 42.3.1-2
Severity: important
In the journal I've discovered a disturbing segfault:
Mai 25 23:43:55 AnonymizedComputerName dbus-daemon[820]: [system] Successfully activated service 'org.freedesktop.realmd'
Mai 25 23:43:55 AnonymizedComputerName systemd[1]: Started Realm and Domain Configuration.
Mai 25 23:43:55 AnonymizedComputerName realmd[1322]: claimed name on bus: org.freedesktop.realmd
Mai 25 23:43:55 AnonymizedComputerName kernel: show_signal_msg: 14 callbacks suppressed
Mai 25 23:43:55 AnonymizedComputerName kernel: gnome-shell[1068]: segfault at 0 ip 0000000000000000 sp 00007fffc9a8c8a8 error 14
Mai 25 23:43:55 AnonymizedComputerName kernel: Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
Unfortunately, dmesg and syslog contain no additional information from the last moments before the segfault. In my understanding, this failure leads to the inability to start Wayland (and xorg kicks in instead).
Who is the real culprit and what to do? Any remedy or bugfix? Any (as small as possible) set of packages to upgrade?
I run Debian stable, kernel 5.10.0-23-amd64, with a few updated packages from testing. Probably related packages: linux-image-amd64 5.10.179-1, realmd 0.16.3-3, systemd 247.3-7+deb11u2, dbus 1.12.24-0+deb11u1.
The severity is raised because the bug is (in my understanding) a major reason for why Wayland is unusable on my machine.
Gratefully,
AlMa

Any time you are seeing a segfault or other crash, a backtrace
<https://wiki.debian.org/HowToGetABacktrace> would be useful information.
systemd-coredump is usually the easiest way to get this:
<https://wiki.debian.org/HowToGetABacktrace#Core_dump>

However, since you are running a mixed stable/testing system, you are on
your own: Debian does not have the resources to support every possible
combination of packages.

As I've said on another of your bug reports, mixing stable and testing is
not a supportable situation, and even if we found a fix for this crash,
if you don't upgrade then you will never receive that fix. Please upgrade
to a fully testing system and see whether the crash persists.

A mixed Debian stable/testing system or a non-up-to-date Debian testing
system are both higher-risk environments than an up-to-date system that is
entirely Debian testing. After you have got onto the "upgrade to testing"
train, you should stay on it until the next station.

Debian testing is currently in hard freeze and is expected to be released
as Debian 12 in mid June, with the same packages that are currently in
testing, so at the moment it is just as tightly-controlled as stable
normally is.

    smcv