Description of problem: Testing Fedora-32-20200221.n.0, system fails to reboot at the end of the installation: ... Running post-installation scripts . Storing configuration files and kickstarts . Installation complete. Press ENTER to quit: Error in atexit._run_exitfuncs: Traceback (most recent call last): File "/usr/lib/python3.8/site-packages/dasbus/client/handler.py", line 474, in _get_method_reply return self._handle_method_error(error) File "/usr/lib/python3.8/site-packages/dasbus/client/handler.py", line 485, in _handle_method_error raise error File "/usr/lib/python3.8/site-packages/dasbus/client/handler.py", line 472, in _get_method_reply result = call(*args, **kwargs) File "/usr/lib/python3.8/site-packages/dasbus/client/handler.py", line 57, in sync_call return connection.call_sync( gi.repository.GLib.Error: g-io-error-quark: The connection is closed (18) Pane is dead (status 0, Fri Feb 21 16:30:42 2020) Version-Release number of selected component (if applicable): anaconda-32.24.1-1.fc32 How reproducible: Everytime Steps to Reproduce: 1. Install using any method, hit enter to reboot.
Created attachment 1664804 [details] syslog
Created attachment 1664805 [details] storage log
Created attachment 1664806 [details] program log
Created attachment 1664807 [details] packaging log
Created attachment 1664808 [details] anaconda log
Created attachment 1664809 [details] dbus log
This is affecting all openQA installer image tests, for 32 and Rawhide. Proposing as a Beta freeze exception at least - I guess we can argue this technically isn't a blocker as the install is complete and if you reboot any other way the installed system should boot, but it obviously looks bad, and it's awkward that it affects openQA (if it doesn't get fixed quickly, I'll have to hack around it).
I think the most likely culprit here is https://github.com/rhinstaller/anaconda/commit/ee2cfcf0d109296c99316ee7669105a71abdf562 , assigning to Vendula...I'll try and test an anaconda build with that reverted...
can't really manage to revert that successfully, or grok what it's doing wrong hard enough to fix it. All the dbus stuff is a bit complex to figure out from outside. sorry!
Fixed in a pull request: https://github.com/rhinstaller/anaconda/pull/2332
Thanks vendula! I'm away till Tuesday afternoon without a laptop so I can't test right away, but fix makes sense to me. If we could get it built for rawhide and f32 asap that'd be great, as it'll unblock a lot of openqa tests.
Discussed during the 2020-02-24 blocker review meeting: [1] The decision to classify this bug as an AcceptedFreezeException was made: "We don’t think the counts as a blocker, as by the time it rears it’s ugly little head, the system is already installed, but it looks bad, is limiting automated test coverage, and its fairly trivial to fix." [1] https://meetbot-raw.fedoraproject.org/fedora-blocker-review/2020-02-24/f32-blocker-review.2020-02-24-17.00.log.txt
Confirmed fixed in Fedora-32-20200225.n.0 , thanks.