ci/tickets

Fork 1

Redefine shared tests to work with large updates #543

New issue

Open

opened 2026-02-09 21:32:13 +00:00 by lecris · 2 comments

lecris commented

2026-02-09 21:32:13 +00:00

Owner

Currently many tests are disabled when the bodhi updates get too big because of:

Jenkins would queue up too many jobs because it runs for each koji build instead of each update
Testing-farm does not recognize bodhi update artifacts
- Upstream PRs submitted for this:
  - https://gitlab.com/testing-farm/gluetool-modules/-/merge_requests/1026
  - https://gitlab.com/testing-farm/nucleus/-/merge_requests/268

Currently many tests are disabled when the bodhi updates get too big because of: - Jenkins would queue up too many jobs because it runs for each koji build instead of each update - Testing-farm does not recognize bodhi update artifacts - Upstream PRs submitted for this: - https://gitlab.com/testing-farm/gluetool-modules/-/merge_requests/1026 - https://gitlab.com/testing-farm/nucleus/-/merge_requests/268

lecris added the

labels

2026-02-09 21:32:13 +00:00

lecris added this to the Board project

2026-02-09 21:32:14 +00:00

adamwill commented

2026-02-09 21:50:42 +00:00

Owner

You'll need to figure out resultsdb reporting and CI messages for reporting against bodhi updates instead of koji builds, I guess? I have prior art for this in openQA of course, let me know if I can help, and try to follow the same formats for everyone's sanity :)

lecris commented

2026-02-09 21:56:56 +00:00

Author

Owner

@adamwill wrote in #543 (comment):

You'll need to figure out resultsdb reporting and CI messages for reporting against bodhi updates instead of koji builds, I guess? I have prior art for this in openQA of course, let me know if I can help, and try to follow the same formats for everyone's sanity :)

Ah indeed. And I do not want to dig into how to do that with Jenkins, unless we can have a simple hack around it. If we can figure it out in rmdepcheck there is some hope, otherwise it would have to wait for packit to catch up on this.

@adamwill wrote in https://forge.fedoraproject.org/ci/tickets/issues/543#issuecomment-376870: > You'll need to figure out resultsdb reporting and CI messages for reporting against bodhi updates instead of koji builds, I guess? I have prior art for this in openQA of course, let me know if I can help, and try to follow the same formats for everyone's sanity :) Ah indeed. And I do not want to dig into how to do that with Jenkins, unless we can have a simple hack around it. If we can figure it out in [rmdepcheck](https://github.com/fedora-ci/rmdepcheck-pipeline/blob/5ef347dc51b20d4d7958067f80e5c28a3f0d77ca/Jenkinsfile#L77-L79) there is some hope, otherwise it would have to wait for packit to catch up on this.

No milestone

No project

No assignees

2 participants

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

ci/tickets#543

No description provided.

Rows
Columns