status of rpc servers in s3, and which process should handle ncalcrpc and ncacn_ip_tcp?

[Andreas Schneider]

On Wednesday 07 March 2012 16:18:14 you wrote:
> On Wed, 2012-03-07 at 14:05 +1100, Andrew Bartlett wrote:
> > On Tue, 2012-03-06 at 10:30 +0100, Andreas Schneider wrote:
> > > On Saturday 03 March 2012 13:41:36 Andrew Bartlett wrote:
> > > > I've been looking more closely at the rpc server setup, and I remain
> > > > quite confused by the current state.
> > > > 
> > > > In particular: Which process should handle ncaclrpc and ncacn_ip_tcp
> > > > connections for embedded RPC services?
> > > 
> > > Hi Andrew,
> > > 
> > > > Currently it looks like it would be a race between the parent smbd,
> > > > spoolssd, lsasd and any forked SMB-handling child, because the events
> > > > are added to the main event loop.
> > > 
> > > it is turned off by default and you shouldn't activate it. smbd isn't
> > > able to handle incoming ncalrpc or tcpip correctly. That's the reason
> > > why we have a preforked spoolssd and lsasd.
> > 
> > On that basis, I have prepared the attached patches.  Currently, having
> > an endpoint mapper implementation causes ncalrpc pipes and TCP sockets
> > to be created for embedded services, but as you say, this isn't going to
> > work.
> > 
> > If we want to make any of these services support ncalrpc and TCP, we
> > will need to move them to a child, like lsasd.
> 
> Updated patches (currently under test) are in my s3-rpc_server branch.

If they pass make test I'm fine with it. The lsa tests are run over tcp and 
ncalrpc too.

Do you want me to sign and push the patches?
 

	-- andreas

-- 
Andreas Schneider                   GPG-ID: F33E3FC6
www.cryptomilk.org                asn at cryptomilk.org