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

Re: Package EVR problems in Fedora 2007-10-31



Ralf Corsepius <rc040203 <at> freenet.de> writes:
> Which is the correct approach, IMO, because testing will receive
> experimental stuff, which is by no means guaranteed to end up in
> "updates".

But surely it should end up in updates-testing or updates for higher versions 
and in Rawhide too (except during the final devel freeze, in which case you 
should prepare the update as a 0-day testing update if it isn't suitable for 
tagging straight into the upcoming release).

IMHO the following inequalities should hold, where "stable" is release + 
(stable) updates:
Fedora n stable <= Fedora n testing
Fedora n stable <= Fedora n+1 stable
Fedora n testing <= Fedora n+1 testing
During the final devel freeze, Fedora n+1 stable/testing for Rawhide are the 
updates pending to be pushed to stable resp. testing at the point of the 
release. Otherwise, Rawhide is both Fedora n+1 stable and testing, which 
implies that Rawhide should normally have newer versions than updates-testing 
in any release when it's not frozen.

        Kevin Kofler


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