Date: Tue, 18 Aug 1998 11:37:36 -0500 (CDT) From: Chris Dillon <cdillon@wolves.k12.mo.us> To: Vadim Belman <voland@plab.ku.dk> Cc: "freebsd-stable@FreeBSD.ORG" <freebsd-stable@FreeBSD.ORG> Subject: Re: Intel EtherExpress Pro 10/100B network collisions Message-ID: <Pine.BSF.3.96.980818110920.7290A-100000@duey.hs.wolves.k12.mo.us> In-Reply-To: <199808181036.MAA28563@plab.ku.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 18 Aug 1998, Vadim Belman wrote: > Hello, > > after having heard so many positive opinions about the > network card in <subj>, we decided to give it a try on > our new FreeBSD 2.2.7 machine (previously we used > 3COM 905 TP for all of ours FreeBSD boxes). > > Surprisingly, we found that Intel board produces a _huge_ > number of network collisions, almost for every output > packet. > > The strange thing is that we anyhow had a terrific > transfer rate - absolutely the same as with 3COMs > (~920 Kb/s on a 10 mbps connection). > > It is definitely a problem with the card, for we have also > tried 3COM on the same machine, with predictable zero number > of collisions. There is quite likely that this is not a problem > with a particular Intel card, for we have also tried another one > --- with the same bad results. > > We have 3COM switches installed in our location. > > Did anyone experience the same or similar problem with > cards in question? > > We'd appreciate any help; we think the results we obtained > are kind of strange. I set up a small network for someone using a P5-200 with a EtherExpress Pro 100/B running a 2.2.5-STABLE copy of FreeBSD (just before 2.2.6 came out, actually) and Samba for SMB filesharing. The hub I used was a small 16-port 10BaseT Netgear job (I hear good things about NetGear.. don't know if they're true or not). My situation is still to this day worse than what you are experiencing. Luckily the load is rather low so the person doesn't care or even notice, but I still want to eventually resolve this. The three clients are running Win95 OSR2.1 with 3COM 3c509s (EtherLink III). Upon transferring any data between the server and the Win95 boxen, there is an interesting collision phenomena. According to netstat, the amount of collisions vs. data flow seem to follow an almost half-duty-cycle sawtooth waveform pattern (you electronics nuts will know what I'm talking about). Actual data transfer rates start out high, and collisions begin to inch up until the collision rate is so high that absolutely no data passes, and then the collision rate suddenly drops to zero and data starts going full-speed again. The cycles continue and each cycle lasts approximately 4 or 5 seconds. The cabling appears to be fine and the same situation occurs even if I remove all nodes from the network except the ones under test, and I have tested each client with the same results. If you ever figure out what is going on or if anyone else could shed light on this I'd also be grateful. :-) P.S. I always figured it was just because FreeBSD was too damn fast for those poor Windows clients and they couldn't choke the data down fast enough. ;-> -- Chris Dillon - cdillon@wolves.k12.mo.us - cdillon@inter-linc.net /* FreeBSD: The fastest and most stable server OS on the planet. For Intel x86 and compatibles (SPARC and Alpha under development) (http://www.freebsd.org) */ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.980818110920.7290A-100000>