rawhide report: 20140113 changes

Adam Williamson awilliam at redhat.com
Tue Jan 14 19:03:30 UTC 2014


On Tue, 2014-01-14 at 11:39 -0700, Michal Jaegermann wrote:
> On Tue, Jan 14, 2014 at 10:10:27AM -0800, Adam Williamson wrote:
> > On Tue, 2014-01-14 at 10:59 -0700, Michal Jaegermann wrote:
> > > 
> > > I am afraid that I miss something.  The above will retire obexfs.
> > 
> > No, it won't 'vanish'. It will still be in the Fedora package
> > repositories. I'm not talking about getting the package removed from
> > *users' systems*. I'm talking about getting it removed from *the Fedora
> > package collection*.
> 
> Now I understand that we are talking about different things. :-)
> I doubt very much that you will be able too rewrite a history that way.
> Copies may be sitting in umpteenth repositories on which you have zero
> control.  Besides that would possibly break something for somebody
> running older Fedora distributions for whatever reasons.
> 
> If you are concerned only about the main repo for the current rawhide
> then I would think naively that it is enough to ask nicely
> correpsponding admins.  Mirrors will sync with that and this will be it.
> Am I mistaken?  But even with an old package physically present - once
> obsoleted it will be not used.  So what is a big deal?

Yes, you're mistaken. I wasn't really asking for uninformed guesses, I
was asking for precise information from the people who actually know
about the relevant processes.

The process here is
https://fedoraproject.org/wiki/How_to_remove_a_package_at_end_of_life .
What I am basically asking is whether one needs to adopt an orphaned
package in order to put it through that process, or whether it is OK for
any packager to file a retirement request on a package that is clearly
an appropriate candidate for requirement. But I phrased it more
generally in case this is a 'known situation' with other wrinkles to it.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net
http://www.happyassassin.net



More information about the test mailing list