releng

#12976 Incorrect FTI bug
Opened by jjames. Modified

Open
Close Issue
  • Describe the issue
    Bug https://bugzilla.redhat.com/show_bug.cgi?id=2399680 was filed a few hours ago. It is wrong. As of this writing, nauty-2.9.0-4.fc43, which provides libnauty-2.9.0.so()(64bit), is still current in F43. An update which transitions to nauty-2.9.1-1.fc43, which provides libnauty-2.9.1.so()(64bit), has been submitted for stable, but has not actually been pushed stable yet: https://bodhi.fedoraproject.org/updates/FEDORA-2025-b85553831e. However, that update also contains gap-pkg-nautytracesinterface-0.3-2.20250620git75cf4e5.fc43, which is linked against libnauty-2.9.1.so()(64bit).

Somehow the FTI script looked inside an update that has been submitted, but not pushed, to stable to find the nauty package update, but failed to notice that the same update contains an updated version of gap-pkg-nautytracesinterface.

  • When do you need this? (YYYY/MM/DD)
    N/A
  • When is this no longer needed or useful? (YYYY/MM/DD)
    N/A
  • If we cannot complete your request, what is the impact?
    Incorrect FTI bugs will continue to be filed.

By the way, the bug report ends with:

P.P.S. If this bug has been reported in the middle of upgrading multiple dependent packages, please consider using side tags: https://docs.fedoraproject.org/en-US/fesco/Updates_Policy/#updating-inter-dependent-packages

A side tag was, in fact, used to build these packages and submit them as a single update.

You used both a side-tag and a buildroot override?
https://bodhi.fedoraproject.org/overrides/nauty-2.9.1-1.fc43

That BRO is likely the cause of the FTI bug, and it doesn't look bogus.

Note also the report mentions:

"P.S. The data was generated solely from koji buildroot, so it might be newer than the latest compose or the content on mirrors. To reproduce, use the koji/local repo only, e.g. in mock:"

so the buildroot override is likely the problem here?

I'm not sure what we can do here?

Metadata Update from @phsmoura:
- Issue tagged with: low-gain, low-trouble, ops

The bug was opened because the package was not installable in the Koji buildroot, which was correct because of https://bodhi.fedoraproject.org/overrides/nauty-2.9.1-1.fc43 -- when it expired and/or the update went stable, the next run of the script closed that bugzilla.

A side tag was, in fact, used to build these packages and submit them as a single update.

Technically, correct. Should the bugzilla say this instead?

If this bug has been reported in the middle of upgrading multiple dependent packages, please consider using side tags, rather than instead of buildroot overrides?

It could also be improved to say:

The data was generated solely from koji buildroot, so it might be newer than the latest compose or the content on mirrors. It also includes buildroot overrides.
Metadata
None

low-gain low-trouble ops

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.