Re: samba run-time modification to smb.conf files

Luke Kenneth Casson Leighton (lkcl@regent.push.net)
Fri, 24 Apr 1998 22:16:49 +0000 (GMT)

Date:	Fri, 24 Apr 1998 22:16:49 +0000 (GMT)
From:	Luke Kenneth Casson Leighton <lkcl@regent.push.net>
To:	Multiple recipients of list <samba-technical@samba.anu.edu.au>, "Christopher R. Hertel" <crh@NTS.Umn.EDU>
Subject: Re: samba run-time modification to smb.conf files
In-Reply-To: <199804242144.QAA20113@unet.unet.umn.edu>

On Sat, 25 Apr 1998, Christopher R. Hertel wrote:

> >
> > there are numerous dce/rpc "set" calls. is it reasonable to support
> > these?
> >
> > if so, we will need to add the capability (currently only found in swat)
> > to smbd to write to smb.conf files. this is one reason why i suggested
> > that swat (or swat v2) go "via" a samba daemon instead of editing smb.conf
> > files direct.
> ^ directly
> --
>
> Luke,
>
> I don't know the specifics of dce/rpc. Is this something we can freely
> implement?

we _are_ implementing it :-) :-)

> I've been talking about putting an interface in front of passwords, WINS
> database, configuration, and other internal data sets. A while ago, you
> suggested dce/rpc for this. I'd like to consider it, but I need more
> info.

dce/rpc is a means to make procedure calls over a network. it specifies
the means by which the data structures are "flattened" and then
reconstructed.

> Note that such a move will take a great deal of time. Not because it's
> hard, but because we'll want to go at it slowly. If we succeed, we'll
> have an interface that can be used to tie in *separate* LDAP, SQL, etc.
> type tools.

ok, this is independent of what the back-end is: this is the communication
mechanism.