From: Herbert Poetzl (herbert_at_13thfloor.at)
Date: Thu 17 Feb 2005 - 21:08:26 GMT
On Thu, Feb 17, 2005 at 03:36:57PM -0500, Stephen Frost wrote:
> * Herbert Poetzl (herbert_at_13thfloor.at) wrote:
> > On Thu, Feb 17, 2005 at 02:34:24PM -0500, Stephen Frost wrote:
> > > Of course, the state of the Debian vserver package is a direct
> > > consequence of linux-vserver "unstable" labeling, which isn't exactly
> > > something we have a whole lot of say over, though I've been bitching
> > > about it for months anyway.
> > there is no 'unstable' in linux-vserver ;)
> > we have:
> > - "stable" which means:
> > the API and ABI will not change in any incompatible
> > way and new features will not be introduced lightly
> > (compare that to 2.2 or 2.4 kernel)
> Which I assume is the '1.2' release set that pretty much everyone I've
> seen recommend against using...
and it will be the 2.0 release for 2.6 kernel ...
> > - "development" which means:
> > this is the place where new features can be found
> > it is intended for testing, evaluation and if you
> > are bold (or want to use the advantage) even production,
> > but do not expect the API and ABI to be changeless
> > or the patches to be well tested ...
> > (compare that to 2.6 and 2.6-rc*)
> I assume this is associated with the 1.9 set of releases, which is what
> pretty much everyone seems to recommend using, and is what most of us
> are running I believe.
and also the (basically discontinued) 1.3 branch for 2.4
> > - "experimental" which means:
> > we added some new stuff or changed something out of
> > the blue, please give it a try in your test setup and
> > let us know what you think/find/discover ...
> > do not use it in production unless you know what you
> > are doing ...
> > (compare that to the bk releases or -mm)
> Near as I can tell these aren't even released as anything, and I'm
> guessing you're talking about ngnet stuff here.
you are wrong here, any release like 188.8.131.52 for example
is an experimental release, of course ngnet stuff and
such addons are considered experimental or highly
experimental too ...
> > > It also won't be in Debian kernel-patch format as I wouldn't
> > > really want people to get the wrong impression about those patches.
> > which would be?
> That they're official Debian packages, of course.
> > > > this has worked before, so I'm pretty sure it is possible
> > > > again, and as I said, I have absolutely no problem with
> > > > a debian kernel patch, if it is maintained and tested ...
> > >
> > > I'd really like to see the official Debian packages updated and uploaded
> > > w/ decent kernel-patch packages but unfortunately we still have
> > > something of a stand off between the current Debian maintainer and
> > > linux-vserver upstream regarding the state of linux-vserver and if it's
> > > 'unstable' or 'stable'.
> > see above ... no idea what debian 'unstable' means
> > (maybe that it is supposed to break?)
> s/unstable/development/ I believe to get the correct lingo for
> linux-vserver. Perhaps it'd be better if it was called unstable, 'cause
> it seems very much like Debian/unstable, perhaps not perfect, but
> nothing ever is and it works damn well for being called unstable.
unstable for me is something which will hold for
some time but then fall apart (usually accompanied
by some fancy effects) like uranium ...
linux-vserver is not supposed to fall apart ;)
> Vserver mailing list
Vserver mailing list