fedora-ci / general

#524 PR CI unuseable due to s390x waits
Opened by churchyard. Modified

Open
Close Issue

Pretty much all time we send a PR to Python, the scratch build times out on s390x.

On Zuul, the job hangs for 6 hours and then reports a non-voting failure.

On non-Zuul, the jobs never report back (they are forever pending) and the Dist git Tets job never starts. This is https://pagure.io/fedora-ci/general/issue/485

This combined makes the CI experience pretty bad.

Example: https://src.fedoraproject.org/rpms/python3.14/pull-request/97

I merged this one: https://github.com/fedora-ci/dist-git-build-pipeline/pull/41

Raises the timeout, hopefully helps a bit

As for Zuul, @lecris any idea where to raise the timeout for it?

@churchyard any idea what is going wiht Fedora infra? that those builds are taking so long, we also saw timeouts in ppc64le builds :(

No idea.

Opened https://pagure.io/fedora-project-config/pull-request/348, hopefully it addresses it.

I have a case where s390x build succeeded, but was reported as failed in PR: https://src.fedoraproject.org/rpms/thunderbird/pull-request/26#comment-283389 .

I have a case where s390x build succeeded, but was reported as failed in PR: https://src.fedoraproject.org/rpms/thunderbird/pull-request/26#comment-283389 .

Network issue. It's a limitation of the current CI runners having to wait for the builds. With packit, it would be easier to make it more event-based and resolve all of these issues
Metadata
None
None
None
None
Static archive. pagure.io was sunset in 2026; this is a read-only snapshot.
Red Hat Logo

Fedora is sponsored by Red Hat. Learn more about the relationship between Red Hat and Fedora »

© Red Hat, Inc. and others.