[jcifs] Just an idea...

[Michael B. Allen]

Fine. I'll change it.

On Sat, 23 Nov 2002 15:12:55 -0600
"Christopher R. Hertel" <crh at ubiqx.mn.org> wrote:

> On Sat, Nov 23, 2002 at 03:14:01PM -0500, Michael B. Allen wrote:
> :
> > > The SessionSetupAndX Request comes from the client and is sent to the
> > > server, so I don't see how it could be the server type.  As you point out,
> > 
> > Why not? This is how the client indicates to the server what type of
> > CIFS implementation is running on the client.
> 
> That's for an integrated system, where the client and server are both 
> built into the OS.
> 
> > > the client cannot know the server type until it gets the Response message.
> > 
> > The jCIFS client can never know what the server running locally is. Not
> > unless it does some monkey business trying to connect to the local host.
> 
> So you are assuming that the field is supposed to represent the type of
> the local, integrated client *and* server system.  That's not what I'm
> assuming.  I'm saying that the field, when sent in the Session Setup
> Request, represents the client type.  There are a lot of systems that run 
> as clients but offer no services.
> 
> > Because we would be lying.
> 
> ...but "foo" is lying too.
> 
> > If jCIFS is running on a machine that has a
> > "Windows 2000 5.0" server running and we advertise that the local server
> > (NativeLanMan) is "jCIFS" the server might think were running the jCIFS
> > server (yet to be written :~) and it ain't so.
> 
> I don't think that the server thinks anything, and even if it does, the 
> "foo" server is lying too.
> 
> The Server side of such a connection isn't going to do anything based on 
> the presumed assumption that the client has a server of type "X" running.
> The information is only valid within the session.  If someone on the 
> server machine tries to connect to services on the system running jCIFS 
> then they will start a new session, get a new Session Setup, and use the 
> server type returned in the Session Setup Response.
> 
> > So there's nothing we can put in there. We have to put something. Might
> > as well be 'foo'.
> 
> ...or 'jCIFS'.
> 
> I have two reasons for wanting this.  One is that it looks better in the 
> examples in my book.  :)
> 
> The other is that people are embedding jCIFS into other products and 
> projects.  It would be nice to identify ourselves.  Since the value is set 
> by a property, they can override us if they want.  The default, though, 
> could be "jCIFS" as easily as it could be "foo".
> 
> Other than that, it's (obviously) a very minor point.  My belief is that
> the only use of the field is as an identifier in error messages on the
> server side.
> 
> Chris -)-----
> 
> -- 
> Samba Team -- http://www.samba.org/     -)-----   Christopher R. Hertel
> jCIFS Team -- http://jcifs.samba.org/   -)-----   ubiqx development, uninq.
> ubiqx Team -- http://www.ubiqx.org/     -)-----   crh at ubiqx.mn.org
> OnLineBook -- http://ubiqx.org/cifs/    -)-----   crh at ubiqx.org
> 


-- 
A  program should be written to model the concepts of the task it
performs rather than the physical world or a process because this
maximizes  the  potential  for it to be applied to tasks that are
conceptually  similar and, more important, to tasks that have not
yet been conceived.