Package: lightdm Version: 1.32.0-6+b2 Severity: important Forwarded: https://github.com/ubuntu/lightdm/issues/435
We are seeing a regression on Debian 13 (Trixie) where LightDM crashes and is restarted by systemd, which resets the active desktop session. Environment: - Debian 13 (Trixie) - lightdm 1.32.0-6+b2 - lightdm-gtk-greeter 2.0.9-1 - kernel: 6.18.5-pwr70plc - libc6: 2.41-12+deb13u2 This setup worked on Bullseye with LightDM 1.26.0. Configuration: - xvnc.socket listens on TCP 5900 - [email protected] runs: /usr/bin/Xvnc -inetd -query localhost -once -geometry 1920x1200 -SecurityTypes=TLSNone - LightDM has XDMCP enabled In our environment this is triggered by Rapid7/security scanning touching TCP 5900 multiple times fast. The incoming connection activates Xvnc via systemd socket activation, and Xvnc then makes a local XDMCP request to localhost:177. LightDM later crashes when that unmanaged XDMCP session times out. Observed result: - lightdm aborts - systemd restarts lightdm - the active desktop session is reset/disrupted Relevant journal output: g_object_unref: assertion 'G_IS_OBJECT (object)' failed double free or corruption (fasttop) Source ID 30 was not found when attempting to remove it lightdm.service: Main process exited, code=killed, status=6/ABRT Relevant /var/log/lightdm/lightdm.log output: [+173.51s] DEBUG: Got Query(authentication_names=[]) from 127.0.0.1:63690 [+173.69s] DEBUG: Got Request(display_number=1 ...) from 127.0.0.1:63690 [+173.69s] DEBUG: Send Accept(session_id=53140 ...) to 127.0.0.1:63690 [+299.74s] DEBUG: Timing out unmanaged session 53140 [+299.74s] CRITICAL: g_object_unref: assertion 'G_IS_OBJECT (object)' failed [+299.74s] CRITICAL: Source ID 30 was not found when attempting to remove it Expected result: LightDM should clean up timed-out unmanaged XDMCP sessions without crashing. This is reproducible in our environment whenever TCP 5900 is touched and socket activation starts Xvnc -query localhost.
