In Fedora-Rawhide-20200430.n.1, neither the KDE live image nor a system installed with KDE from the network installer manages to boot to a working desktop. You get stuck at a blank screen with background, or at the login screen. 20200428.n.0 booted OK, so I compared the logs. The error happens shortly after startplasma-x11 is called. On 20200428.n.0 we see this: klauncher[1401]: Connecting to deprecated signal QDbusConnectionInterface::serviceOwnerChanged(QString,QString,QString) plasma_session[1424]: org.kde.plasma.session: Starting autostart service "/etc/xdg/autostart/xembedsniproxy.desktop" ("/usr/bin/xembedsniproxy") etc. etc. In the same place in 20200430.n.1, we get this instead: klauncher[1452]: Connecting to deprecated signal QDbusConnectionInterface::serviceOwnerChanged(QString,QString,QString) kdeinit5[1458]: Module "kcm_access" does not actually have a kcminit function startplasma-x11[1374]: "/usr/bin/plasma_session" () exited with code 127 startplasma-x11[1374]: QProcess: Destroyed while process ("ksplashqml") is still running. Proposing as an F33 Beta blocker, violates Basic criterion "A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility." - https://fedoraproject.org/wiki/Basic_Release_Criteria#Expected_installed_system_boot_behavior
Well, I thought this was different because there's really no trace in the logs, but I guess it's *possible* the cause here is the same as https://bugzilla.redhat.com/show_bug.cgi?id=1830380 . I say that because this broke on the same day, we do have pipewire-libpulse in the 20200430.n.1 image but pulseaudio-libs-glib2 in the 20200428.n.0 image, and the KDE upgrade tests did *not* fail in the same compose - which is kinda consistent with the pipewire bug, because the bad pipewire-libpulse probably doesn't get installed on upgrade, only in fresh installs. We can wait and see what happens with the next compose - I sent a pipewire build that should avoid the bug, so if both Workstation and KDE start working again in the next compose, it was very likely the same bug.
KDE *did* start working again, so let's go ahead and assume this was a dupe. *** This bug has been marked as a duplicate of bug 1830380 ***