os-autoinst-distri-opensuse
os-autoinst-distri-opensuse copied to clipboard
os-autoinst
Workaround for bsc#1222411
- This PR mainly does below. Now with these, in OSD, kermit can finish SL Micro 6 bare metal installation successfully.
- usb selection on vh082: exclude usb with key word
Virtual_NetFile - installed os persistent boot setting for kermit: kermit can't set the newly installed
sl-securebootas first boot, so do it in combustion - add code to pass bootparam to firstboot
- recover default selinux config by removing customized config file
- workaround https://bugzilla.suse.com/show_bug.cgi?id=1222411#c9
- usb selection on vh082: exclude usb with key word
The 2 OSD machines selected, kermit and scooter , have been fully enabled and updated in openqa worker settings to do SL Micro 6 test. See https://gitlab.suse.de/openqa/salt-pillars-openqa/-/merge_requests/775. But for general sles15sp6 test, please wait for Julie to finish uefi sles verifications in poo#153037. After https://gitlab.suse.de/openqa/salt-pillars-openqa/-/merge_requests/775 is merged, kermit can serve official SL Micro 6 test directly. Scooter needs to wait until Julie's ticket work is done.
- Related ticket: https://progress.opensuse.org/issues/151498
- Verification run: https://openqa.suse.de/tests/14014368 : PASS
Great PR! Please pay attention to the following items before merging:
Files matching lib/**.pm:
- [ ] Consider adding or extending unit tests in t/
This is an automatically generated QA checklist based on modified files.
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
Could you help show what needle
pass-bootparam-to-firstbootlooks like ? And how the test proceed if it is found ???I saw it. It is here. https://openqa.suse.de/tests/14014368#step/bootloader_uefi/26
This new rd.kiwi.install.pass.bootparam is bit longer, there is possibility it spans two lines. See:
Could you help show what needle
pass-bootparam-to-firstbootlooks like ? And how the test proceed if it is found ???I saw it. It is here. https://openqa.suse.de/tests/14014368#step/bootloader_uefi/26
This new rd.kiwi.install.pass.bootparam is bit longer, there is possibility it spans two lines. See:
Looks like a new needle is needed.
The openqa worker kermit and scooter MR for usb install, is merged.
Merging this PR, will make OSD tests workable in development group. Can we merge this PR now? Please reply before EOB today if anything more needed here.
Could you help show what needle
pass-bootparam-to-firstbootlooks like ? And how the test proceed if it is found ???I saw it. It is here. https://openqa.suse.de/tests/14014368#step/bootloader_uefi/26
This new rd.kiwi.install.pass.bootparam is bit longer, there is possibility it spans two lines.
Looks like a new needle is needed.
A new needle can not solve the problem because this line is too long and its position is unpredictable.
This new rd.kiwi.install.pass.bootparam is bit longer, there is possibility it spans two lines.
Looks like a new needle is needed.
A new needle can not solve the problem because this line is too long and its position is unpredictable.
This parameter is defined by kiwi. We can't make it shorter. When it spans two lines, adding needle with two parts selected per the two lines, shall be fine, I think. Its position can differ for vh015 and kermit. But catching needles for each of the machine shall be fine, right? We do not need to have the test run on every ipmi machine, but only selected machines. And for each machine, it shall be fixed needle based on my observations. I thought about why needles of this differ. I think it is because we can't set the video when booting from iso, which is different from pxe boot when we can set fixed value.
I think merging this pull request is OK at the moment.
There was a new build yesterday in OSD. Our tests can run with this PR merged.