Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!daver!leadsv!oetl1!alan
From: alan@oetl1.scf.lmsc.lockheed.com (Alan Strassberg)
Newsgroups: comp.dcom.lans
Subject: Re: Vance Morrison's PCBridge
Message-ID: <1991Apr01.165051.4039@oetl1.scf.lmsc.lockheed.com>
Date: 1 Apr 91 16:50:51 GMT
References: <91086.144020P85025@BARILVM.BITNET>
Organization: LMSC Rocket Ranch, Santa Cruz, CA.
Lines: 29

In article <91086.144020P85025@BARILVM.BITNET> P85025@BARILVM.BITNET (Doron Shikmoni) writes:
>Has anyone been successful in running PCBridge (on a PC/XT with
>two WD8003E cards)? I'm having the weirdest problems and would
>be interested in hearing from anyone who's using PCBridge in
>production.

	Here's a couple gotcha's we discovered. Initially
	PCBridge was swamped due to broadcast traffic. Code changes
	to incorporate the following have been forwared to Vance.
	'Smart' filtering solved our problem.

	- The only thing that was "broke" was that it left the interrupts
	off when the queue filled up.

	- The "smarter filtering" stuff: it passes Novell broadcasts
	and ARP broadcast packets that have a value other than 0 or
	255 in the low byte of the target IP.  Any other broadcast is
	dropped.  The 0 and 255 values are from various beasts stupidly
	searching for their net masks and some legitimate traffic for
	routers and such.  From netwatch it seemed that this was a large
	part of the traffic.  Since it didn't seem like we should need any
	of this, I throwed it out and it quieted things down real nice.

				alan
-- 

-- 
Alan Strassberg             alan@oetl1.scf.lmsc.lockheed.com
(408) 425-6139              alan@leadsv.ese.lmsc.lockheed.com