[ On Fri, March  3, 1995 at 07:38:18 (-0800), Don Lewis wrote: ]
> Subject: Re: rdist vs. sup
>
> Over a serial link using error correcting modems (V.42, etc.) I'd say
> the most likely packet corruption would be dropped bytes at either end
> of the connection.  This should be pretty detectable by the IP checksum.

Exactly -- the carrier link should be clean, given that the modems are
working correctly.  Now assuming the modem-host cables are short, and
shielded, this connection should be clean too.  The most likely problem
will be in the host's async interface, and I know for a fact that the
NetBSD-1.0/i386 driver will get overrun errors on many systems, even
with 16550's.  The damn i386/i486 is just too slow at processing
interrupts, and the driver has just too much code in it.

> corrupt IP packets.  I'd say it would be rare indeed for a corrupt packet
> to pass the PPP CRC and the IP checksum, since these use different algorithms.

Indeed.

Any time there are huge numbers of IP packet errors there's something
wrong.  I do agree that there are many types of errors which the current
version of TCP will not detect, but it should never come to that!

-- 
						Greg A. Woods

+1 416 443-1734			VE3TCP		robohack!woods
Planix, Inc. <woods@planix.com>; UniForum Canada <woods@uniforum.ca>