[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Proposal to do regular jenkins updates via jessie-updates (Was: Re: Removing Jenkins from Jessie)



On Wed, 2015-04-08 at 23:33 +0200, Niels Thykier wrote:
> On 2015-04-08 22:45, Miguel Landaeta wrote:
> > On Wed, 08 Apr 2015 18:17:59 +0200, Niels Thykier escribió:
> >> [...]
> >>
> >> I had a chat with James Page and Emmanuel Bourg about Jenkins over IRC.
> >>  We concluded that it was infeasible for Debian to maintain Jenkins due
> >> to the lack of upstream commitment to a LTS release-cycle of sufficient
> >> length to match the length of Jessie[1].
> > 
> > Do you think is feasible or acceptable to maintain Jenkins in
> > jessie-updates suite instead?
> > 
> 
> I am not entirely convinced that Jenkins applies to stable-updates
> criteria[1].  However, I am leaving the final call on that to the SRMs.

As someone who was involved in the initial setup of stable-updates, I'm
afraid that I'm not convinced either.

Packages such as clamav get updated to new upstream versions via
stable-updates, but that's mostly because the (anti-)malware landscape
changes sufficiently quickly that it's often not feasible to make small
updates to the existing version in order to remain viable and we serve
our users better by making newer engines available to them. Apologies if
I'm missing something, but that really doesn't seem to be the case for
Jenkins.

https://wiki.jenkins-ci.org/display/JENKINS/LTS+Release+Line suggests
that "long-term" means "supported for three months". I'm struggling to
combine those two ideas, particularly in the context of a Debian stable
release. (Similarly ""battle-tested" — meaning those commits that have
already been a part of a main line release for more than a week".)

I do wonder whether backports might be suitable, but I can't and won't
speak on behalf of the backports team.

Regards,

Adam


Reply to: