Fedora Board Recap 2008-FEB-12

[Josh Boyer]

On Thu, 14 Feb 2008 20:10:52 -0800
John Poelstra <poelstra at redhat.com> wrote:
 
> === Post-release updates of custom spins (2008-01-29) ===
>   * Should the board have to approve them?
>   * We will hosts as many spins as we have space for

Still wondering if there is some kind of hierarchy or priority for
spins.

>   * Need to determine the hosting requirements and limits
>   * How long will spins stay around?

I suggested EOL along with the base they are on.  Any thoughts on that?

>   * http://fedoraproject.org/wiki/RahulSundaram/SpinsProcess
>   * ACTION: Jef to review Rahul's proposal and report back to board
>   * OWNER: Jef Spaleta
>   * '''FOLLOWUP on 2008-02-12'''
>     * Jef is waiting for feedback from Jeremy Katz on release 
> engineering's perspective
>       * Reference: 
> http://fedoraproject.org/wiki/ReleaseEngineering/Meetings/2008-feb-11
>     * People are still not clear on exactly what is required to create 
> an official "Fedora Spin"
>     * Need a clear list of guidelines of what a spin owner is 
> responsible for and what they are required to test
>       * Hoping a test  will come from Jeremy Katz as part of feedback 
> from release engineering

I think this should have said "...test matrix.." ?

>     * We are only talking about spins that use GA packages, thus testing 
> for GA should have given us enough comfort that risk is minimized
>        * We only need to be concerned with new combination of packages 
> that a spin would present

This isn't quite true.  Spins are done against the repos that exists on
the day they are generated.  So if there are issues/bugs in the
base+updates repos on that day, there will be bugs in the spin.  For
example, the recent Xfce spin we sort of lucked out that it pulled in a
kernel with the security fixes for the vmsplice vulnerability.

(Not quite sure what you meant by "GA", so we may be saying similar
things.)

josh