On 10/28/2010 10:59 AM, Connie Sieh wrote:
> On Thu, 28 Oct 2010, Dennis Gregorovic wrote:
>> On Thu, 2010-10-28 at 10:03 -0400, Chris Lumens wrote:
>>>> Ok just to clarify.. I know that this change will not affect CentOS
>>>> 4.9, 5.[6-9] or 6.X.. but would only affect 7.X but will this affect
>>>> RHEL7 if it turns out to need multi-ISO or do we want to wait til
>>> This change will only affect F15 and any distributions derived from it
>>> or later Fedora releases. So, no RHEL .x.
>>> - Chris
>> It's hard to predict the RHEL 7 requirements at this point.
>> Since Fedora itself is not producing split media, even if we kept the
>> code in anaconda, it would not get any use in the Fedora space. So, I'm
> Then use it in Fedora.
> So what is the relationship of Fedora to future RHEL?
It's mostly a temporal relationship
. Yes, RHEL 7 will come from a
branched (future) Fedora. If we absolutely must have split media in RHEL
7, we can put something back in. But the installer team (and presumably
release engineering) are both going to argue against it quite strongly.
There's also not a lot of point in keeping the current code - mostly it
will just atrophy and bit-rot between now and then.
>> inclined to say yank the code and we can cross the RHEL 7 bridge when we
>> get to it.
> So yank it now and put it back, not very effcient. Better to leave it
> in and lets users that need it use it in Fedora, then it will be
> there for RHEL 7 .
There's no real reason to believe code we don't use that's untested for
several years will still work reasonably well. In fact, there's plenty of
reason not to. Remember that when we're talking about RHEL 7, that's the
fairly distant future. It's an age in terms of code. Also the real
efficiency here is simple: we're going to try hard not to put it back.
Anaconda-devel-list mailing list