On Sun, May 28, 2006 at 02:02:38PM -0400, Joey Hess wrote: > It's been a rather long time since the linux-2.6 in testing has gotten > updated. I count something like 24 security holes[1] in the version in > testing that would be fixed if we could get a release in from unstable. > We also cannot really release a new version of d-i until the new kernel > goes in (unless we go back and base it on an old kernel and not include > amd64 -- yugh). > Of the RC bugs: > #368544, #368667 > Only bugs on the version in experimental, although britney doesn't > realize that. Tagged experimental now, so britney should have no problem understanding now. > #349354 > The sarge upgrade issue. However, AFAICS 2.6.15 in testing also > has the same issues so this shouldn't block a new version from > testing. Right -- may require manual overriding due to britney's current lack of version tracking support. > #365455 > Support for some minor powerpc subarches is lost in the new version > of the kernel. Presumably this includes bug #359025. Given all the > other reasons for updating it, the words "omlette" and "eggs" come > to mind. :-/ <nod> > Other issues: > > * NEW > Quite a lot of the arches have the newest version of linux-2.6 > stuck in NEW for a week. I assume the ftpmasters will fix this soon. Seems to be resolved now. > * missing hppa build > The build started on May 22nd, I assume it went missing. According to <http://buildd.debian.org/fetch.php?pkg=linux-2.6&arch=hppa&ver=2.6.16-14&stamp=1148321402&file=log>, this is a build failure. There's also an unrelated build failure on mipsel. It's not *completely* insane to push 2.6.16 in without these two archs, but I'd rather avoid the complications this causes later and just get a proper package into unstable now that builds where it should. > * release frequency > The kernel team tends to release new versions more frequently > than their urgency could let them into testing (average time > between uploads for 2.6.16 is on the order of 4 days). We either > need to get very lucky on timing or they need to decide to put off > uploading the latest and greatest release long enough for the > previous release to get to testing. This is certainly something the kernel team needs to be aware of; cc:ed. -- 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