On 29/09/01 Michael Bramer wrote:
> On Fri, Sep 28, 2001 at 03:05:07PM -0800, Ethan Benson wrote:
> > On Fri, Sep 28, 2001 at 04:25:36PM +0200, Christian Kurz wrote:
> > > No, please reread my previous mails: I claim that people will abandon
> > > debian, if we only support chrooting bind via mount --bind and not
> > > offering alternative at least for users of kernel 2.2.x.
> > yes that is absurd.
> > lets look at this:
> > people see debian only supports chrooted bind on 2.4 kernels because
> > of mount --bind, they can do either:
> > a) manually configure a 2.2 capable chroot for bind manually.
> > b) be irrational, dump debian, and configure a chrooted bind manually.
> > now what exactly did switching distros buy them?
> IMHO bind should support chroot per mount --bind and if Christian use
> 2.2 and he like chrooded bind, he will post a patch to
> submit@bugs.debian.org. Or not Christian?
No. If the bind debian package will only support chroot by using mount
--bind, I will not send in a patch. Especially since like previously
stated in this discussion it's fairly easy to chroot bind this days and
I don't see any need for depending on kernel 2.4.x features like 2.4.x
for this purpose. So instead of sending in a patch, I would stop
installing debian on any machine that should be a nameserver and switch
to an os, where either I get a chrooted bind by default or directly can
build a chroot manually.
Christian
P.S.: I wonder what the hell all you people favoring mount --bind would
do if that option wouldn't be available. Would you then stop chrooting
bind until you get one day such a option or instead try to figure out an
other solution to chroot bind? If the latter is true, then why do we
need to discuss so much about this instead of offering both solutions to
chroot bind and letting the adminstrator decide which he wants to use?
--
Debian Developer (http://www.debian.org)
1024/26CC7853 31E6 A8CA 68FC 284F 7D16 63EC A9E6 67FF 26CC 7853
Attachment:
pgpNq4ppDVLQO.pgp
Description: PGP signature