-----BEGIN PGP SIGNED MESSAGE-----
Susanne Oberhauser wrote:
Pascal Bleser <pascal.bleser(a)opensuse.org>
Dominique Leuenberger wrote:
On Mon, 2008-11-10 at 12:12 +0100, Andreas Jaeger
Note: novfs is now part of our kernel tree:
pm -qlp kernel-default-126.96.36.199-1.1.i586.rpm |grep novfs
So, no need anymore for the kmp.
It IS part of the kernel, but (at least on my
system... I did not update
yet to the 188.8.131.52 kernel) the version is older than what is shipped in
the 2.0 beta client (2.0.0-440 vs 2.0.0-504 from the client)
Ouch, so now it means
that there is no way to upgrade the KMP as it
will conflict with the kernel package :(
ko modules from KMPs go to a different location
(/lib/modules/$kernel/updates) so they don't conflict.
Eh, right, sorry :)
IMHO it would
be better to have it as a KMP that is always built
against the latest kernel in oss-updates, especially if the release
I tend to agree to that anyway, because of the different release cycle
could if they have the quality we need and get released at the
> right time - and the last is one problem, we do not want to touch
> the official repos after we released the goldmaster,
Why not just have them in
a repository of their own ? It would be
non-oss, with EULA/disclaimer/copyright/whatever, but in its own
non-oss repos on build.opensuse.org
is an interesting idea.
You could alternatively ask for a build.novell.com
software, to have a clear and obvious separation of the two.
Indeed, it must be separated, and personally, I really don't bother
whether it's in the OBS or anywhere else.
But not having a proper repository with up-to-date packages for the most
recent openSUSE versions is somewhat ridiculous, when you think about it.
It does sound
quite hopeless, tried in the past already. Seems that
the Novell Client for Linux is developed by the part of Novell that
doesn't care at all about openSUSE.
'not at all' is too strong -- it's about priorities and about risk:
Really. I suppose you had a look at this bug already:
I opened that bug 2007-04-07 (even though it was already an issue before
that), it was closed for whatever reason (because there's a public beta
for 10.3 and 11.0 that isn't even in a repository ?) so IMHO it still
isn't quit there yet.
I agree, "not at all" might be too strong, "barely" or "not
probably more adequate.
How can it not be a priority to have Novell's own Novell Client on
Novell's own community distribution ?
I understand that having it built, tested and made available on SLED and
Windows is the top priority, but really, I'm sure I'm not the only one
who finds that not having a working Novell Client on openSUSE is.. let's
say somewhat ironical at best :)
openSUSE has no support or QOS commitments whatsoever,
and no API
commitments, nothing. It's a world in its own and there is a lot of
goodwill but nobody would buy a great car if you *maybe* get it fixed,
which is if someone happens to be interested in fixing it.
While I understand the point you're trying to make as far as comparing
openSUSE to SLE*, that does sound quite extreme.
"No support commitments whatsoever" ? Except 2 years of support
regarding security patches and major bugfixes.
No API commitments, probably, although there are policies in place wrt a
best effort API+ABI stability, e.g. providing security and bug fixes as
backports instead of bumping to upstream versions.
If the openSUSE distribution was about "maybe get it fixed", then none
of us would be here right now.
So if Novell officially commits to support a product
on top of
openSUSE, and some bug that shows in the product actually needs a fix
in the base OS, then the users are bad off.
That doesn't sound like a probable use case though, I'd rather expect
that the Novell Client would require fixes because of bugs in it, or
that it would require changing a few bits in its novfs KMP if the
distribution requires a newer kernel for security reasons and the FS API
in the kernel differs (not all that likely though).
That's why. It's actually a "we do
care": we care about the
"we do care" == making the Novell Client for openSUSE a very low
priority job ?
I'm not sure :)
I agree, though, a 'beta' release with
goodwill commitments could be a
Just give it to us, we'll test and report bugs, that'll benefit
everyone. But also take care of fixing the bugs that are reported and
put a somewhat higher priority on this. That would mean "we do care" to
-o) Pascal Bleser <pascal.bleser(a)opensuse.org>
-- I took the green pill
_\_v FOSDEM::7+8 Feb 2009, Brussels, http://fosdem.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
-----END PGP SIGNATURE-----
To unsubscribe, e-mail: opensuse-project+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-project+help(a)opensuse.org