On Tue, 19 Jul 2016 11:30:30 +0200 Thomas Moschny <thomas.moschny@xxxxxxxxx> wrote: > 2016-07-18 23:57 GMT+02:00 Kevin Fenzi <kevin@xxxxxxxxx>: > > I could tell you more about your specific updates if you list > > them... > > This one took a bit more than 6 days from submission to testing: > https://bodhi.fedoraproject.org/updates/FEDORA-2016-e5b5fbfa86 Yep. This is one of the other cases where something happened. :) So, when we first started doing atomic updates in bodhi composes we ran into some problems with f22-updates-testing. Some were server (selinux, etc), some were bodhi, some were rpm-ostree. In debugging and trying to fix that we added a side repo with a newer rpm-ostree. This allowed us to interate quickly without having to officially build and submit rpm-ostree updates. We got everything composing and moved on. Last week however we were adding the f24-updates and f24-updates-testing atomic updates and when doing so noticed we still had the old side repo enabled for f22-updates-testing. This was removed/disabled. However, things broke again there (only in f22-updates-testing, which your update above was in). Making things more difficult, many releng folks were on vacation that week (I was). Finally we got it fixed up by unpushing a rpm-ostree update that was sitting in f22-updates-testing for a year, re-enabling the side repo and doing a compose, then disabling it again. This took some of my saturday morning (to see what was going on), then discussion in the releng meeting monday morning to come up with a plan to fix things up, then the actual fix monday. kevin
Attachment:
pgpmWowemwftY.pgp
Description: OpenPGP digital signature
-- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://lists.fedoraproject.org/admin/lists/devel@xxxxxxxxxxxxxxxxxxxxxxx
