Bugzilla – Bug 1226919
[Build 20240622] [openQA] Not rebooting completely after installation
Last modified: 2024-06-28 18:42:48 UTC
## Observation openSUSE Tumbleweed for s390x can not be restarted for initial boot after the package installation. openQA test in scenario opensuse-Tumbleweed-DVD-s390x-autoyast_zvm@s390x-zVM-vswitch-l2 fails in [installation](https://openqa.opensuse.org/tests/4295521/modules/installation/steps/28) The autoinst log by Yast is saying the following: [2024-06-24T13:44:47.720851Z] [debug] [pid:10802] ||| finished installation autoyast (runtime: 1665 s) [2024-06-24T13:44:47.720924Z] [debug] [pid:10802] ||| post fail hooks runtime: 35 s [2024-06-24T13:44:47.722998Z] [debug] [pid:10802] stopping overall test execution after a fatal test failure [2024-06-24T13:44:47.741138Z] [debug] [pid:10783] stopping autotest process 10802 [2024-06-24T13:44:47.758987Z] [debug] [pid:10783] [autotest] process exited: 0 [2024-06-24T13:44:47.859496Z] [debug] [pid:10783] done with autotest process [2024-06-24T13:44:47.860272Z] [debug] [pid:10783] stopping command server 10793 because test execution ended [2024-06-24T13:44:47.860454Z] [debug] [pid:10783] isotovideo: informing websocket clients before stopping command server: http://127.0.0.1:21033/7f872Tpz1cx7wHok/broadcast [2024-06-24T13:44:47.878367Z] [debug] [pid:10783] commands process exited: 0 [2024-06-24T13:44:47.978739Z] [debug] [pid:10783] done with command server [2024-06-24T13:44:47.978888Z] [debug] [pid:10783] isotovideo done [2024-06-24T13:44:47.979564Z] [debug] [pid:10783] backend shutdown state: -1 [2024-06-24T13:44:47.980320Z] [debug] [pid:10803] Passing remaining frames to the video encoder [2024-06-24T13:44:48.015279Z] [debug] [pid:10803] Waiting for video encoder to finalize the video [2024-06-24T13:44:48.015388Z] [debug] [pid:10803] The built-in video encoder (pid 11183) terminated [2024-06-24T13:44:48.015564Z] [debug] [pid:10803] sending magic and exit [2024-06-24T13:44:48.015788Z] [debug] [pid:10783] received magic close XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":41809" after 28293 requests (28160 known processed) with 0 events remaining. [2024-06-24T13:44:48.059401Z] [debug] [pid:10783] backend process exited: 0 XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":41885" after 18910 requests (18910 known processed) with 0 events remaining. xterm: fatal IO error 11 (Resource temporarily unavailable) or KillClient on X server ":41809" [2024-06-24T13:44:48.060981Z] [info] [pid:10783] ::: backend::driver::_collect_orphan: Driver backend collected unknown process with pid 11433 and exit status: 1 [2024-06-24T13:44:48.061235Z] [info] [pid:10783] ::: backend::driver::_collect_orphan: Driver backend collected unknown process with pid 11436 and exit status: 0 [2024-06-24T13:44:48.061586Z] [info] [pid:10783] ::: backend::driver::_collect_orphan: Driver backend collected unknown process with pid 11250 and exit status: 1 [2024-06-24T13:44:48.061922Z] [info] [pid:10783] ::: backend::driver::_collect_orphan: Driver backend collected unknown process with pid 11434 and exit status: 84 [2024-06-24T13:44:48.062233Z] [debug] [pid:10783] stopping backend process 10803 [2024-06-24T13:44:48.062310Z] [debug] [pid:10783] done with backend process 10783: EXIT 0 ## Test suite description Create HDD for s390x textmode ## Reproducible Fails since (at least) Build [20240622](https://openqa.opensuse.org/tests/4292726) ## Expected result Last good: [20240617](https://openqa.opensuse.org/tests/4286238) (or more recent) ## Further details Always latest result in this scenario: [latest](https://openqa.opensuse.org/tests/latest?arch=s390x&distri=opensuse&flavor=DVD&machine=s390x-zVM-vswitch-l2&test=autoyast_zvm&version=Tumbleweed)
That was a not matching needle.
It is failing continuously.
There is no y2logs tarball here at this bug (where it should be to ensure it won't disappear as OpenQA clear away old tests). There is no y2logs tarball at the test. There is no test description here. There is no test description at the test. All I can see at the test are some screenshots where I can only guess that it stops some time after booting. Since it complains about unprivileged users not being able to log in yet, this appears to be stuck some time before the multiuser systemd target. But there is a login prompt, so some getty process is already active at the console.
https://openqa.opensuse.org/tests/4295521/logfile?filename=serial0.txt openSUSE Tumbleweed Installation o3zvm003 login: So it already knows its hostname which probably means that it got beyond the DHCP and DNS stage.
The messages in comment #0 about video and X11 failing seem to indicate that this is a bug either with kernel video drivers or X11 or both. Full messages: https://openqa.opensuse.org/tests/4295521/logfile?filename=autoinst-log.txt And the test uploads the video.ogv and some more files, but not an y2logs tarball despite complaining that the test failed. Why?
Sarah, please, read https://en.opensuse.org/openSUSE:How_to_Write_a_Good_Bugreport Following it should us with understanding bugreports from others. Thanks in advance.
This is not a YaST installer bug. There are zero indications that this might be one. The most likely candidate is a kernel video driver problem, but with only the little information that this test gives us I am not optimistic that the kernel maintainers can do very much here.
I just saw that the complaints about video seem to be about the video encoder for the video.ogv; but I am not 100% sure. But X11 terminates with >> XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":41885"
And for future bugs, please choose a useful bug subject. That auto-generated "openQA test fails in installation" is completely nondescript and even misleading: It does NOT fail in installation; the installation went fine. It was rebooting competely AFTER the installation that failed.
Sorry, HuHa! I did not create bugreports via openQA for a long time, and it is difficult for us to receive more information than the uploaded logs and the openQA output based on the fact that we do not have direct access to a mainframe.
Stefan, I offer you a free beer in Kater Murr today (Pre Conference Party) or an invitation to our SUSE/openSUSE BBQ Party this week because of my mistake. You are allowed to choose.
That was a bug withing the Tumbleweed release and is fixed now.