[Samba] domain-free multi-user use cases

[Eric Levy]

I have browsed documentation and previously engaged this group, trying
to resolve a means toward a deployment with specific characteristics.

Through such investigation, it has become apparent to me that the set
of use cases presently supported by Samba generally fall into one of
the following two classes:

   1. Single-user mounts of a server share on a client, without support
      from any third node.
   2. Mounts of a server share on a client, often multiuser, with support
      from a domain server or equivalent kind of node.

I might easily call these two classes of use case the single-user cases
(1) and the domain-server cases (2).

Two observations about this dichotomy are striking. One is that in
practical use, many desired configurations will fall easily into one of
these classes. The other observation is the ease with which someone may
imagine a valuable use case not within either class.

The many combinations of needs for various deployments indicate a wide
gap between these two classes. I might suggest this gap represents a
third class, which I might call the domain-free multi-user cases. With
the convergence of technologies, and with deployments entering
increasingly diverse environments, it may seem that these use cases are
becoming increasingly important.

In my earlier conversation in this group, I described my needs as
follows:

   What I want is multiple users on the client accessing the same mount
   but with different permissions enforced for each. I want the
   permissions to reflect the permissions for the corresponding users
   on the NAS.

   It seems by now it has been made clear that it is impossible to
   achieve this result without some kind of domain server...

Support for such functionality without a domain server would require
development of new components on the client and server to handle
certain functions currently available only through the domain server.
Such augmentation would call for good design, but is doubtless
feasible, at least in principle.

Naturally, one of the nice things about being Microsoft is that after
you sell licenses for a few clients and a file server, you also can
sell a license for a domain server. The profit-focused designs of
software system offered by Microsoft and other companies might
represent a more narrow range of options than is useful generally. The
Samba project might benefit from designs expanded to serve a more
inclusive variety of user needs, with fewer constraints inherited from
commercial models. 

I realize that new support of the kind I am describing is a substantial
undertaking. I am interested in learning how these thoughts might be
received considering the current state of development and future
ambitions.