From owner-freebsd-hardware Fri Sep 19 06:45:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id GAA12667 for hardware-outgoing; Fri, 19 Sep 1997 06:45:34 -0700 (PDT) Received: from plains.NoDak.edu (tinguely@plains.NoDak.edu [134.129.111.64]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id GAA12660 for ; Fri, 19 Sep 1997 06:45:32 -0700 (PDT) Received: (from tinguely@localhost) by plains.NoDak.edu (8.8.5/8.8.5) id IAA03070; Fri, 19 Sep 1997 08:45:20 -0500 (CDT) Date: Fri, 19 Sep 1997 08:45:20 -0500 (CDT) From: Mark Tinguely Message-Id: <199709191345.IAA03070@plains.NoDak.edu> To: hardware@FreeBSD.ORG, sysseh@workcover.qld.gov.au Subject: Re: ATM cards for FreeBSD in Australia Sender: owner-freebsd-hardware@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I'm not too clued in about ATM standards in Australia, so I have no > idea if any of the U.S. cards work here. Someone I know in Canberra wishes to > snoop ATM packets using the BPF device, which if the drivers follow the usual > network model, should be in place already. Does anyone have answers to the > following? > don't listen to those telling you can only snoop PDU local to the machine. there are two options of viewing other PDUs. 1) snooper ports in switches, redirect packets going down other lines 2) optical taps. these are T connection that you attach to a pair of fibers. there is a paritially reflective mirror inside that reflects a small portion of the light to you ATM card. ------->------- | | ==================>========+===== | -- to ATM card 1 SWITCH1 SWITCH2 ==================<========+===== | -- to ATM card 2 | | ------>-------- if you want to take advantage of the ATM card's SAR, it would be nice to know ahead of time what AAL of the VC you are looking at. At worse case, you can capture raw cells and then manually do the SAR functions. > Are there any ATM cards that are supported by FreeBSD current at all? the Fore and Efficient cards are supported. I wrote a IDT NICStAR driver that has BPF hooks, a graduate student wrote tcpdump parsing print routine for LANE (that work was done on a Sun), I am looking at doing the Classic IP over ATM print routine for the IDT (the HARP stack does IP over ATM). > Are they likely to work in a box running the SMP code? I think the capture into memory/disk, then post-process (like a commerical Sniffer does) is your best bet to keep from losing information. --mark. PS. you may be interested in asking ATM questions to the freebsd-atm mailing list.