From owner-freebsd-mobile Mon Aug 27 11:34:19 2001 Delivered-To: freebsd-mobile@freebsd.org Received: from ptavv.es.net (ptavv.es.net [198.128.4.29]) by hub.freebsd.org (Postfix) with ESMTP id EBBCD37B407 for ; Mon, 27 Aug 2001 11:34:16 -0700 (PDT) (envelope-from oberman@ptavv.es.net) Received: from ptavv.es.net (localhost [127.0.0.1]) by ptavv.es.net (8.10.1/8.10.1) with ESMTP id f7RIYER21957; Mon, 27 Aug 2001 11:34:14 -0700 (PDT) Message-Id: <200108271834.f7RIYER21957@ptavv.es.net> To: "Chad R. Larson" Cc: reader@newsguy.com, freebsd-mobile@FreeBSD.ORG Subject: Re: Frustrating network problem - need diagnotic help In-reply-to: Your message of "Mon, 27 Aug 2001 10:35:49 PDT." <20010827103549.B475@freeway.dcfinc.com> Date: Mon, 27 Aug 2001 11:34:14 -0700 From: "Kevin Oberman" Sender: owner-freebsd-mobile@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > Date: Mon, 27 Aug 2001 10:35:49 -0700 > From: "Chad R. Larson" > > On Mon, Aug 27, 2001 at 10:07:03AM -0700, Kevin Oberman wrote: > > I'm afraid not. I do have my 802.3 handy and the source address field > > is not interpreted at all. The I/G bit (individual/group) bit is > > ALWAYS interpreted in a destination address, so setting the "locally > > admin" bit does not over-ride this functionality. The import is that > > many devices (including switches) make forwarding decisions based on > > the setting of this bit. > > We did a bunch of multicasting stuff at my prior job, where we were > building high-availability clusters of Pyramids. At the time, the > O/S didn't support IP aliases on the interfaces, so we set up > locally administered multicast addresses for the machines so clients > could find their application even if it had migrated within the > cluster. > > We had to put static ARP entries in the Cisco routers, as they were > pretty snooty about dealing with multicast at all. > > > In any case, a hardware MAC address that has the group address bit set > > is clearly broken. > > Unless it is a deliberate multicast. In this case, not likely. But > not intrinsically broken. I will stand by my statement that a HARDWARE MAC with the group address bit set is broken. It violates the spec and delivering a device with a multicast source address is broken even if the spec was not violated. R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: oberman@es.net Phone: +1 510 486-8634 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-mobile" in the body of the message