Ancestors

Written by Jo Shields on 2025-01-31 at 18:59

It's a "being gaslit by the CI server" kind of day

=> More informations about this toot | More toots from directhex@tech.lgbt

Toot

Written by Jo Shields on 2025-01-31 at 19:40

I think GitLab wanted vengeance against me for not having reason to complain for nearly 2 weeks. I've got 2 weeks of pent up CI misbehaviour in one inexplicable bug to try and figure out.

=> More informations about this toot | More toots from directhex@tech.lgbt

Descendants

Written by Jo Shields on 2025-01-31 at 19:57

The issue describing GL's failure on this was opened FIVE YEARS AGO.

"SLO Missed" indeed.

=> More informations about this toot | More toots from directhex@tech.lgbt

Written by Jo Shields on 2025-01-31 at 20:21

• if: '$CI_COMMIT_BRANCH != $CI_DEFAULT_BRANCH'

when: manual

WHY DOES THIS DISABLE ALLLLLL OTHER UNRELATED JOBS IN A PIPELINE

=> More informations about this toot | More toots from directhex@tech.lgbt

Written by Ludovic Henry on 2025-02-01 at 21:17

@directhex I can feel the frustration from this side of the pond

=> More informations about this toot | More toots from ludovic_dev@mastodon.social

Written by Ben Hutchings on 2025-02-01 at 23:33

@directhex It's not because of implicit dependencies between pipeline stages, is it?

=> More informations about this toot | More toots from bwh@tech.lgbt

Written by Jo Shields on 2025-02-02 at 00:05

@bwh as best i can determine, it’s a broken interaction between “this is a MR build” and “this is a branch build” of the same commit. What’s particularly fishy is the MR build fully ignores a strict needs: on one of the jobs it’s decided to suppress, and allows the new manual job to just ignore it silently

=> More informations about this toot | More toots from directhex@tech.lgbt