[Oneiric-Topic] SRU Process

Chuck Short chuck.short at canonical.com
Thu Mar 31 17:45:21 UTC 2011


On Wed, 30 Mar 2011 11:25:44 -0400
Etienne Goyer <etienne.goyer at canonical.com> wrote:

> On 11-03-30 11:05 AM, Scott Kitterman wrote:
> > On Wednesday, March 30, 2011 10:40:36 AM Chuck Short wrote:
> >> Hi,
> >>
> >> I do not have the statistics in front of me, but I believe most of
> >> users are using LTS releases of Ubuntu. The policy of cherrypicking
> >> fixes from the development releases does not scale in my opinon. We
> >> should offer PPAs for users who want to use a new version of for
> >> example Apache. Or go through the list of packages we support and
> >> see if we can get it to qualify as a micro release update.
> > 
> > We can also do a lot of this through backports.  We are very close
> > to having the backports only install packages that users explicitly
> > request from backports (just waiting on an LP change that's in
> > progress), so it will be much safer to use going forward.
> 
> The mechanism itself is really only half the question.  I am more
> interested in the level of commitment we are willing to make in
> keeping certain key software "fresh" in LTS.  Whether we deliver
> these in backports, PPA or some other mechanism is really just an
> implementation detail, IMHO.
> 
> 

Right when I brought this up, I was more interested in the burden of
tracking down the fixes in upstream code, backporing the fix, and
asking for user testing. If you look at:

http://people.canonical.com/~chucks/SRUTracker/sru-tracker-bugs.html

We have a lot of requests for SRU bug fixes which takes time to
fix. Depending on the workload that what have, we do what we can. 

From the systems administrator perspective if they want to use newer
versions of apache, mysql, etc then great they can use backports/ppa.
If they want to be a more cautious then they can still use the
*-updates pocket. Its really all about choice in this case.

chuck




More information about the ubuntu-server mailing list