#4317 Koji reports scratch buildd as a successl despite fedpkg srpm would fail
Closed: Invalid by tkopecek. Opened by ahajkova.

When there's a mistake in a spec file changelog date, fedpkg srpm or fepkg build done locally fails. If Merge request with such a bogus spec file is submitted, Fedora CI - scratch build will be a success. I think it's very misleading and scratch build should be marked as failed in a case there's such a warning in build.log:

RPM build warnings:
bogus date in %changelog: ...

An example of such a case:
https://src.fedoraproject.org/rpms/gdb/c/73a92e455cb369f39adf2bb2e0cb3e52eec44ef5?branch=rawhide

https://kojipkgs.fedoraproject.org//work/tasks/8494/128938494/build.log


rpmbuild just issues the warnings. It doesn't fail neither refuse to create srpm. Maybe you've some rpmbuild hooks (e.g. linter) locally?

$ fedpkg clone -a gdb
Cloning into 'gdb'...
remote: Enumerating objects: 11650, done.
remote: Counting objects: 100% (307/307), done.
remote: Compressing objects: 100% (93/93), done.
remote: Total 11650 (delta 277), reused 227 (delta 212), pack-reused 11343
Receiving objects: 100% (11650/11650), 7.80 MiB | 1.29 MiB/s, done.
Resolving deltas: 100% (7861/7861), done.
$ cd gdb/
$ fedpkg srpm
Downloading gdb-16.2.tar.xz from https://src.fedoraproject.org/repo/pkgs
######################################################################## 100.0%
setting SOURCE_DATE_EPOCH=1739145600
Wrote: /XXXX/gdb/gdb-16.2-2.fc43.src.rpm

Metadata Update from @tkopecek:
- Custom field Size adjusted to None

Metadata Update from @tkopecek:
- Issue close_status updated to: Invalid
- Issue status updated to: Closed (was: Open)

This issue has been migrated to Fedora Forge:
https://forge.fedoraproject.org/koji/koji/issues/4317

Please continue any further discussion there.

Metadata