In message <20031020211540.5aa4e73c.davem@redhat.com>,"David S. Miller" writes: >Nobody uses ipv4 multicast. Do you know how I can tell? The >fact that I _NEVER_ see bug reports for ipv4 multicast usage >yet I know there must be bugs lurking there. actually i did submit a bug report with the ipv4 multicast not too long ago. we actually try to use ipv4 multicast to distribute time via ntp instead of lots of p2p connections (on a lane/atm network, using multicast to distribute time consumes far fewer vc's since each host doesnt have a private connection to the time servers). this has been somewhat 'flaky' at times and i am not sure who to blame (ntpd or kernel). i cant see a reason not to have reliable multicast in the kernel. if someone wants to write it and support it (and the code is 'well written') should the code be denied entry simply because it has a small user base? (that would be bad news for atm of course). - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
