On Fri, 2013-10-11 at 04:42 +0200, Kevin Kofler wrote: > Jerry James wrote: > > On Aug 13, update FEDORA-2013-14530 acquired enough karma to be > > autopushed to stable. It went stable on Aug 15. > > > > The first update, FEDORA-2013-14567, stayed in limbo for awhile until > > positive karma was given to it on Sep 28 and 29, causing it to reach > > its karma threshold on Sep 29, and be autopushed to stable. On Sep > > 30, it went stable, wiping out the -2 build. > > The real issue there is autokarma. I have complained several times about how > broken that concept is. If the decision to push to stable had been made by a > sentient being, chances are this would not have happened. A sentient being made a decision to use autokarma. Maintainers can choose not to. The 'real issue' is clearly, as the previous posters stated, a bug in Bodhi. There are probably several; I suspect there's a different problem in the case where there's one update with multiple packages and then someone submits a newer update containing a newer build of only *one* of those packages. I don't think Bodhi does anything that could be considered 'the right thing' there either. Given the complexity of this case, I do think it would be a good idea to have a backstop 'sanity check' in Bodhi which would never allow a superseded update to be pushed stable without an explicit manual override. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin DOT net http://www.happyassassin.net -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
