SUMMARY *** NOTE: If you are reporting a crash, please try to attach a backtrace with debug symbols. See https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports *** STEPS TO REPRODUCE 1. Connect to a wifi without internet access. 2. Wait about 2-3 min. 3. Do anything with Plasma (eg. open application launcher, open an applet, etc.) OBSERVED RESULT Plasmashell freezes for random amounts of time EXPECTED RESULT Plasmashell continues to work normally. SOFTWARE/OS VERSIONS Windows: macOS: Linux/KDE Plasma: 5.24.3 (available in About System) KDE Plasma Version: 5.24.3 KDE Frameworks Version: 5.90.0 Qt Version: 5.15.3 ADDITIONAL INFORMATION ROG Zephyrus G14 running Fedora 36 with the latest stable kernel (5.17). Observed same result with kernel 5.16. To fix the issue, the network needs to get disconnected. If one doesn't have a terminal open or can use a Terminal, the user needs to wait for Plasma to unfreeze and then hurry to disconnect from the connection. It happens on both Wayland and X11
Forgot to add a SUMMUARY - sorry!!!, so here is it: Whenever you connect to a connection (using applet, settings or automatically) without actual connection to the Internet, Plasmashell freezes. I just had a internet-outage and had this happen to me. I tried restarting my PC, restarting plasmashell using kquitapp5 and kstart5. Neither worked.
I can confirm this (with 4G LTE USB modem / not wifi). I have limited bandwidth per day. So when the internet quota runs out, and I click disconnect (from plasma-nm), then more often than not the plasma freezes. If i disconnect the (no quota) internet from modem control panel (via browser 192.168.0.1) then no freeze. If i just unplug the usb modem from usb port, then no freeze. Maybe it's related to dns resolver, because with dnscrypt+dnsmasq the freeze time is much shorter compared to systemd-resolved. I don't see any clue from journalctl or journalctl -f. OS: Archlinux KDE Plasma: 5.24.5 KDE frameworks: 5.93.0 Qt: 5.15.3
Coincident or someone has fixed it really fast? Because I don't have this (comment 2) problem anymore since I wrote a comment here. For my use case (comment 2), it is fixed. Thanks to all who contributed. Hopefully all issues related to this bug are also solved. I still don't know what cause the problem & what fix the problem, though.
*** This bug has been marked as a duplicate of bug 406110 ***