Date: Fri, 11 Apr 2003 14:25:40 +0700 From: Max Khon <fjoe@iclub.nsu.ru> To: Motonori Shindo <mshindo@mshindo.net> Cc: freebsd-net@freebsd.org Subject: Re: Bug in ARP requests Message-ID: <20030411142540.A14117@iclub.nsu.ru> In-Reply-To: <20030411.161056.71083042.mshindo@mshindo.net>; from mshindo@mshindo.net on Fri, Apr 11, 2003 at 04:10:56PM %2B0900 References: <3E942121.7A3647EB@pipeline.ch> <20030411.161056.71083042.mshindo@mshindo.net>
next in thread | previous in thread | raw e-mail | index | archive | help
hi, there! On Fri, Apr 11, 2003 at 04:10:56PM +0900, Motonori Shindo wrote: > From: Andre Oppermann <oppermann@pipeline.ch> > Subject: Bug in ARP requests > Date: Wed, 09 Apr 2003 15:33:21 +0200 > > > I got hit by this pretty hard when I connected a FreeBSD/Zebra box > > to AMSIX. Appearently the Cisco and Juniper boxes don't answer to > > broken ARP requests if the target hardware address field is not set > > to NULL but filled with random memory junk. While debugging this we > > got really confused by the tcpdumps... Ethereal saved the day because > > it has a much nicer display than tcpdump. > > I also came accross this problem when I was testing the equipment > (wireless access point, in fact) made by the company I currently work > for. At that time, I thought that it was specific my company's > equipment, but now I'm amezed to know that there are so many > networking equipments that don't accept such a bogus ARP Request. > > BTW, tcpdump prints out the target hardware address in ARP Request if > it isn't all zero and that made me realized that the problem was > caused by FreeBSD's ARP request. tcpdump saved a day for me:-) ok, rev 1.97 MFC'ed a few minutes ago /fjoe
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030411142540.A14117>