Thanks for the feedback, Doug.  It's much appreciated.

DN>The server program simply gets the max number of clients (let's call it X)
DN>from it's config file or the command-line. Creates the named pipe on the
DN>server specifying X connections. You then create X number of threads which
DN>each do a DosConnectNPipe and then what for a client. You then route msgs

Doesn't OS/2 place a limit on the number of active threads?  If so, and
given that some people might want to run very large BBS'es on an OS/2
LAN (100+ nodes), couldn't there be some problems with this approach?
I'm not aware that there's any reason that a 100-thread server process
is *impossible*, but isn't that a log of resources to hog for one door?

Forgive me if this is another dumb question... I make no claims to
expertise here.   It sounds as if what I'd need to do to avoid
piggishness (at very least) is have one "listener" thread and then spawn
a new thread for each actual "hit" (rather than spawning MaxNodes
threads at startup).

Again, thanks for the info!

Joel

 * OLX 2.1 TD * Forget stealing.  It's the cat juggler that bothers me.

--- WILDMAIL!/WC v4.12