[971] in linux-net channel archive
Re: telnet/ftp/rlogin/rsh not working between Linux machines. (fwd)
daemon@ATHENA.MIT.EDU (Malcolm Beattie)
Wed Aug 23 20:03:46 1995
From: Malcolm Beattie <mbeattie@sable.ox.ac.uk>
To: shaver@neon.ingenia.com (Mike Shaver)
Date: Wed, 23 Aug 1995 10:05:21 +0000 (BST)
Cc: linux-net@vger.rutgers.edu
In-Reply-To: <199508221030.GAA00259@neon.ingenia.com> from "Mike Shaver" at Aug 22, 95 06:30:37 am
Mike Shaver writes:
>
> Arnt Gulbrandsen mumbled something vague about:
> > > Or you could add a hear() system call that would return a struct
> > > something *, which could be checked for stuff like source and
> > > destination address and port before calling accept()...
> >
> > Not sufficient. When you've checked it, what can you do? accept()
> > now or later. getpeername() on a select()able socket can do what you
> > suggest hear() should, but a reject() call is absolutely necessary.
>
> I'd actually want two: reject() sends the ICMP/RST packets and such,
> and ignore() just clears the entry from the queue without a word to
> the other machine.
>
> But yeah, hear() alone isn't enough.
>
> I wonder what the TLI mechanisms look like... anyone know if *BSD has
> TLI, or if I can get the source anywhere else?
I've never really used TLI, but according to the "OSF/1 Network
Applications Progamming Guide" sitting next to me, you apparently
do the following. t_listen waits synchronously (by default) for
connection requests and when one arrives, it gives you the information
about the remote end-point. You can then either do t_accept which
accepts the connection or do t_snddis which does an abortive
disconnect (presumably meaning that it does RST without sending
the SYN/ACK).
--Malcolm
--
Malcolm Beattie <mbeattie@sable.ox.ac.uk>
Unix Systems Programmer
Oxford University Computing Services
