smb://

[Christopher R. Hertel]

> On Wed, Dec 27, 2000 at 10:13:05PM -0600, Christopher R. Hertel wrote:
> > I assume that we will only use smb:// to connect to shares and not other 
> > services, so there will be no need to specify a NetBIOS name suffix.  
> 
> Interesting but not programmatically practical as I do not believe all
> names registered by a host are replicated by the browse list which is
> where the information that would be displayed by the workgroup# semantic
> would be obtained.

All relevent information would be there.  That is, all SMB services 
should be listed (eventually, assuming proper configuration) by the 
browse list.

> It would require broadcasting node status requests
> and then *trying* to capture the packet storm that followed. Get 10
> people doing that on a regular basis and the Admins would descend.

I don't follow.

> But with such functionality you might list Microsoft Exchange MTA with
> something like:
> 
> smb://workgroup#87#

But this is not an SMB service, so using it in conjunction with smb:// 
would not make sense.

> or find where the user jdoe is logged in in anticipation of sending them
> a win pop-up with:
> 
> smb://jdoe#03#

Hmmm... I don't know if win pop-up is carried over smb.  I *think* it is. 
Still, as you point out user names are not listed in the browse list.  I
seem to recall, also, that some versions of Windows (98? ME?) don't
register the <user>#03 names at all.  I think that this puts win pop-up
out of scope. 

> ...etc. Also there is really no garantee that the service advertised
> is really functional. For example W2K advertises some kind of Exchange
> support by default yet I don't think that means you can send such a host
> e-mail directly.

If true, this again suggests that we limit the smb:// URI to accessing 
the browse list and actual SMB services.

> Along these lines one might more accurately target services by allowing
> the Service field of the tree connect to be specified like:
> 
> smb://server|LPT1

The browse list does identify the type of service (printer vs. file 
share).  I don't think that this is necessary.

> or something but in practice I have a feeling this would be a total mess
> not to mention you pushing the limits of how much you can obfuscate your
> documentation :~)

Yep.

Chris -)-----

-- 
Christopher R. Hertel -)-----                   University of Minnesota
crh at nts.umn.edu              Networking and Telecommunications Services

    Ideals are like stars; you will not succeed in touching them
    with your hands...you choose them as your guides, and following
    them you will reach your destiny.  --Carl Schultz