I think MS just did us (and themselves) a disservice.

Kevin Whitney Kevin.Whitney at sun.com
Mon Jan 29 16:18:14 GMT 2001


	
	Perhaps I haven't folled the full thread, but
	please check again.  I believe the idea is that
	existing connections may be terminated *only if*
	the new connection from an existing client comes 
	in with the VC number of '0'.

	Kevin Whitney


"Michael B. Allen" wrote:
> 
> On Tue, Jan 09, 2001 at 11:20:26AM -0500, David Collier-Brown wrote:
> > MS article Q106211 says:
> > ---
> >    Windows NT does not allow you to make multiple connections to a
> >    shared network server from the same workstation if you attempt to
> >    use more than one set of credentials. If you attempt to make two
> >    or more connections to the same server using two or more sets of
> >    credentials, you will receive the following error message:
> 
> The CIFS specification goes as far as to say:
> 
> 2.5.1       Connection Management
> 
> Once a connection is established, the rules for reliable transport
> connection dissolution are:
> 
> o If a server receives a transport establishment request from a client
>   with which it is already conversing, the server may terminate all
>   other transport connections to that client.  This is to recover from
>   the situation where the client was suddenly rebooted and was unable
>   to cleanly terminate its resource sharing activities with the server.
> 
> But NT and friends don't actually do this AFAIK. Not yet anyway. When
> writing jcifs I took nothing for granted in that all sessions are
> multiplexed over a single transport and all trees are multiplexed over
> any suitable session and so on. The only problem is when the users are
> in different Java VMs. I guess it depends on what their definition of
> "a client with which it is already conversing" is. If it's the calling
> netbios name then it's not a problem because you can dynamically generate
> a name. Richard's libsmbclient does this and jcifs generates names like
> JCIFS34_151_A3. If thier definition is based on host or ip then there
> would be a problem but I seriously doubt MS could truly realize such a
> goal because like us, they too have dependancies.
> 
> Mike
> 
> --
> signature pending




More information about the samba-technical mailing list