coordination #130582: [epic] Upgrade all our infrastructure, e.g. o3+osd workers+webui and production workloads, to openSUSE Leap 15.5
Upgrade OBS package CI checks to Leap 15.5 (os-autoinst+openQA) size:M
To use the most recently released Leap 15.5 as CI container base in os-autoinst and openQA we should adapt that and also run the according OBS checks on the same version of Leap
- AC1: Leap 15.5 is used as CI container base in os-autoinst+openQA
- AC2: OBS checks run on pull requests for both os-autoinst and openQA test on Leap 15.5
- Find out where this is configured and update to use Leap 15.5 both for os-autoinst and openQA, see https://openbuildservice.org/help/manuals/obs-user-guide/cha.obs.scm_ci_workflow_integration.html#sec.obs.obs_scm_ci_workflow_integration.use_cases.test_build_pull_request just update https://github.com/os-autoinst/os-autoinst/blob/master/.obs/workflows.yml or maybe even extend to have 15.4+15.5. If it's easier just replace the version. If it's possible test both
- Verify that it works, e.g. by opening a pull request for both os-autoinst and openQA
Although it looks at first glance as if osd-deployment is failing because we're hard-coding 15.4 what I think is actually happening here is that the
deploy.env picked up $OS_VERSION from osd which is still on 15.4.
Nevermind me. There's some outage and if anything salt got upset because the build was still in progress at the time. Nothing directly related.
OBS checks should be covered by https://github.com/os-autoinst/openQA/pull/5197 and https://github.com/os-autoinst/os-autoinst/pull/2325
For some reason builds are excluded: https://build.opensuse.org/package/show/devel:openQA:GitHub:os-autoinst:openQA:PR-5203/openQA
That's why we don't get any results back to PRs anymore
- Status changed from Workable to In Progress
- Assignee set to jbaier_cz
I would guess it could be connected with the "service" OBS service outage (see the slack discussion from comment 6), it looks good in https://github.com/os-autoinst/openQA/pull/5199. It looks to me, that some OBS magic responsible for expanding the
_branch_request was not happening.