Spin Updates Policy Draft

[Jeremy Katz]

On Tue, 2008-03-11 at 12:48 -0800, Jeff Spaleta wrote:
> On Tue, Mar 11, 2008 at 12:32 PM, Jeremy Katz <katzj at redhat.com> wrote:
> >  The actual release of a *release spin* involves the full brunt of
> >  release testing.  Some of the release criteria cover it (not entirely)
> >  and so we do a fair bit of testing to ensure that things like installing
> >  still works including some of the variety around that.
> 
> Can the testing be described in enough detail so that a peer group can do this?
> If manpower is the underlying concern, I'm not against blocking on
> this until additional community manpower can be found to do it. But
> releng has to be okay with community doing the testing, blocking until
> installs works and then unblocking again.  

It's not releng that needs to be okay with it (well, to some extent) --
it's more QA.  So wwoods is your man.

> If this proposal goes
> through we're already going to be blocking on the establishment of a
> peer review to even get access to the trademarks.  I would expect that
> its in that group's best interest to do the testing necessary for
> install issues to even get access to the trademarks.

Doing the testing is one thing when you're talking about the initial
release...  I'll put money on corners being cut on updates.  Not to
mention that there are some live images that are "definitive" from a
Fedora perspective, ie the ones that are all but guaranteed to be there
from release to release.  Having updates of a random spin (let's say FEL
just for the sake of argument) but not of the ones that we _more_ push
people to is very odd looking.  And the driving force behind the more
"definitive" spins isn't a SIG.  It's the release process.  And the
people pushing the release process get a little busy pushing the next
one ;-)

> >  The other thing that is almost guaranteed to be a problem in an "update"
> >  is getting things to fit.  If it's a CD sized live image, then it's a
> >  constant struggle to fit.
> 
> I tried to cover this. If the maintainers can't keep the update under
> the size limit of the target media original release.. full stop.

Then I guess I really don't have to worry since it's never actually
going to be a concern.  From one day to the next I can see size
fluctuations of 10-20 megs :-)

Jeremy