home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!ucbvax!LUDWIG.CTD.ORNL.GOV!LPZ
- From: LPZ@LUDWIG.CTD.ORNL.GOV ("Lawrence MacIntyre - 615.576.0824")
- Newsgroups: comp.os.vms
- Subject: Re: Handling multiple client connections with VMS TCP/IP servers?
- Message-ID: <930104120624.21c002ab@LUDWIG.CTD.ORNL.GOV>
- Date: 4 Jan 93 17:06:24 GMT
- Sender: daemon@ucbvax.BERKELEY.EDU
- Distribution: world
- Organization: The Internet
- Lines: 30
-
- >Using UCX2.0, VAXC3.2 under VMS5.5-2.
- >
- >When writing VMS TCP/IP servers, is there an elegant method of having VMS
- >sub/det processes handle multiple client connections, a'la U**x fork[/exec]?
- >
- >Is multi-threading the server (QIOs and ASTs) the only method? In which case
- >how are telnet, rsh, ... connections, that create network processes under a
- >different UIC, handled?
- >
- >Questions, questions, and so little documentation. The usual apologies (FAQ,
- >etc.) With thanks in advance. Mark Daniel.
-
- Mark:
-
- I can't reply to you because you didn't include an address and the mailer
- didn't leave one that works. There is an elegant way of handling client
- connections, and you already noticed it. Multithreading the server is it!
- Check out sys$examples:db_server.c. This is a DECnet server, but the concepts
- are the same, just substitute TCP code for the DECnet code. Unix fork/exec
- isn't elegant, it's sloppy and lazy *unless* you have a multiprocessor machine
- to run the forked processes on. All it does is make unix programmers think
- that forking is the one true way to write code.
-
- Lawrence
- ~
- -----------------------------------------------------------------------------
-
- Internet: MACINTYRELP@ORNL.GOV US-Snail: Lawrence MacIntyre
- LPZ@ORNL.GOV AT&T: 615.576.0824
-
-