On Sunday 27 January 2008, Kevin Kofler wrote: > One more thing: you're quick to blame the security team approval process > when it delays your Fedora 8 update, This is not about any particular update, and I don't know why you're pointing fingers back at me about something different. I saw something that smelled like a broken process and tried to provide as accurate an example as possible to illustrate my observations, hence used the xine-lib case with what I experienced it with, hoping to get feedback from those who have designed it and are applying it saying whether it works as intended (and if, why). I also asked for instructions in case there was something I should have done differently. > but this is already the third update > you're pushing to Fedora 7 updates-testing, Ok, I'll bite. The first one went to testing because in addition to a security fix it was a version bump from version 1.1.7 to 1.1.9.1. I now think this was a mistake and I should not have touched F-7 at all. The second one (trivial non-security 1.1.9+ regression fixes) went also to testing because nobody had notified me whether the previous testing update worked or not. The 3rd one was an update to 1.1.10 which contained a security fix and some other pretty harmless looking changes - I decided to push that directly to stable because of the nature of those changes and more importantly because meanwhile a confirmation comment arrived that the latest 1.1.9.1 incarnation worked for some people. Bodhi turned that into the 3rd testing request. At the time of filing the 3rd request (more precisely a bit before that) I also revoked the existing 1.1.9.1 testing->stable update request because I had no idea I wouldn't be able to push the new one directly to stable and thought it'd take the same time for the 1.1.9.1 testing->stable to be processed as the 1.1.10 directly to stable one. > and you appear not to have requested a push to stable for any. Yes, I have. I filed that request immediately after the first comment arrived in Bodhi that someone had tested the F-7 update and found it working (thanks, Rex!). > Many maintainers don't even test their NON-security updates on all Fedora > versions before they push them. (Hey, you're lucky if they even tested it > on ANY distro. ;-) ) You may think that's a bad idea, VERY much so, and I will not participate in that madness, but that's a rant for another day. > but at least for > security updates, I think getting it out quickly is more important. For easily reviewable security fix updates only, agreed. -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
