Microsoft cooperating with the Samba Team?

[Luke Kenneth Casson Leighton]

On Thu, 5 Aug 1999, Christopher R. Hertel wrote:

> I think the key line is:
> 
> > > translated:
> > > There is a general problem with Samba
>     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> 
> A bit of fud.
> 
> I know that Luke has done some digging into W2K, but as far as I know
> we're not going to worry about it until it becomes a market force.  People
> are playing with it now, but very few are "using" it. 

the interesting thing is that some people are reporting nt5 beta / samba
compatibility problems to microsoft, as a bug in nt5.

i have been tracking the results of improvements to microsoft's
implementation of msrpc through various betas / hot-fixes / service packs
for eighteen months, now.  sp3, sp4 rc1.40, sp4 rc1.99, sp4, sp5 beta, nt5
beta 1 and beta 2.  it's quite fascinating to watch, particularly as it
looks like microsoft is using nt5 to test [much more robust / stringent]
MSRPC code that later goes into an nt4 service pack.

various sections of the samba implementation have had to be reworked to
deal with these improvements, which are usually tightening up of the
marshalling code.  as these improvements continue, we see more and more
MSRPC "fault" PDUs come back to various MSRPC calls that were not well
understood at the time, but it worked, and it didn't break anything, so it
went into cvs.  the browsing call (NetShareEnum?) is a good example.

microsoft's job of improving the MSRPC code is _probably_ not over, yet.
until all services have been tackled (i.e equivalent functionality
implemented in samba, including things like exchange), we cannot tell.
it's going to take a long time :-)

luke