On Mon, May 08, 2006 at 03:04:16PM -0400, Nathanael Nerode wrote: > These all have RC bugs and are 'leaf' packages in etch. > (As such, they will all get back in easily if the RC bugs are fixed.) > Personally, I'd kick the lot out of testing, but you can pick and choose. > Rationale. > * It would make a substantial and easy dent in the RC bug count for etch, > increasing the average package quality for etch immediately. > * There seems to be little incentive for packagers to fix preexisting RC bugs > if their packages have already reached testing, due to the general policy > that preexisitng RC bugs are ignored for testing propagation. While this > is certainly a necessary policy for packages with substantial reverse > dependencies, it doesn't seem particularly appropriate for leaf packages. Except that even leaf packages become an issue in the case of library transitions. If we have to add a hint to either force in or remove a leaf package tied to a lib transition that has a pre-existing RC bug in testing, that's busywork for the release team. Yes, maintainers need to take RC bugs seriously, and should expect their packages to be removed from testing (and ultimately, the release) if they don't tend to them. Unfortunately, the current rate at which we're removing such packages isn't making much headway against the stream of incoming bugs; though what's really depressing is the number of RC bugs in unstable-only packages... > This list so far only covers packages beginning with "A": amounting to > nine source packages. > # 361139 (security) > remove acidbase/1.2.2-1 > # 359065, 364550 > remove ept/1.90.1 > # 363030 > remove aegis/4.21-2 > # 358342 > # I believe the plan was to drop 2.4 kernels for etch, and these are > # strictly 2.4 modules (and for i386 only, too; and 2.6 works very consistently > # in i386 at least.) Perhaps these should be removed from unstable too? > remove alsa-modules-i386/1.0.10+1 > # 360713 -- patch unapplied by maintainer, should be NMUed probably > remove and/1.2.1-2 > # 365199 > remove araneida/0.90.1-3 > # 327564 > remove aria/1.0.0-13 Hints added. > # 266407, 321771: this does have a reverse recommends from sgml2x > remove alcovebook-sgml/0.1.2-7 It also has a reverse-build-depend of dh-kpatches. Please propose an alternative fix for that build-dependency first. > # 364651, 366042 in unstable > # reverse recommends from kde-extras, reverse suggests from ion3-scripts > remove amarok/1.3.8-1 A rather popular package according to popcon, and the RC bugs against it don't seem reproducible and may be specific to a backend. I'm holding off on removing this for the moment. Thanks, -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. vorlon@debian.org http://www.debian.org/
Attachment:
signature.asc
Description: Digital signature