Date: Sat, 13 Nov 2004 14:19:18 +0200 From: John Hay <jhay@icomtek.csir.co.za> To: freebsd-net@freebsd.org Subject: multicast socket behaviour on IPv4 and IPv6 Message-ID: <20041113121918.GA59462@zibbi.icomtek.csir.co.za>
next in thread | raw e-mail | index | archive | help
Hi Guys, While debugging a misbehaving program, I found that on FreeBSD, if you bind() a udp socket to a multicast address, so that you can receive on it, and you then mistakenly use it to send, the behaviour between IPv4 and IPv6 differ. On IPv6 you get a error EOPNOTSUPP, which is understandable, but on IPv4 the packet is actually send with the source address being the multicast address. Is that the expected behaviour? Looking through the code, I see that in IPv6 there is a lot of sanity checks in the output routine, but very little to check the source address in the IPv4 output routine. John -- John Hay -- John.Hay@icomtek.csir.co.za / jhay@FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041113121918.GA59462>