Recently there was an outage of dist-git causing some of the tickets that were processed at that time to end in inconsistent state.
To prevent this we should first check the availability of external systems (dist-git, Bugzilla, PDC) and then process the request or put it back to queue.
I just had this reoccur when requesting a branch:
https://pagure.io/releng/fedora-scm-requests/issue/50745
in this case, just retrying the request works, luckily, but the problem is more severe when requesting a new repo
Metadata Update from @zlopez: - Issue assigned to jnsamyak
Metadata Update from @jnsamyak: - Assignee reset
Metadata Update from @zlopez: - Issue tagged with: cpe
Metadata Update from @zlopez: - Issue assigned to zlopez
I want to make the whole scm_request_processor more reliable, so the whole process could be retried when something happens and just change what is not correctly set yet.
scm_request_processor
The change is done in https://pagure.io/fedora-infra/toddlers/pull-request/337#
Metadata Update from @zlopez: - Issue status updated to: Closed (was: Open)