[jboss-jira] [JBoss JIRA] Commented: (JGRP-515) UDP.java requires the multicast address in the multicastsocket constructor to fix crosstalk on certain operating systems

[ http://jira.jboss.com/jira/browse/JGRP-515?page=comments#action_12364655 ] Bela Ban commented on JGRP-515: ------------------------------- Well I couldn't sleep without checking on that. Seems like the kernel guy is right and linux doesn't suck so much :) I've hard coded mcreceive to bind to 228.8.8.8 instead of any this way: /* construct a multicast address structure */ memset(&mc_addr, 0, sizeof(mc_addr)); mc_addr.sin_family = AF_INET; mc_addr.sin_addr.s_addr = htonl(3825731592); /* this is int representation of 228.8.8.8 */ mc_addr.sin_port = htons(mc_port); After the change everything seems to be working fine: 1. 2 processes can bind to the same multicast address/join the same multicast group 2. process will not receive any multicast messages to other groups Interesting was to see a process binding to 228.8.8.8 and joining 228.8.8.9. I don't know how does java do it but it has to bind to the proper address as well join the multicast group and not just bind to any address. I hope they didn't do it so that once bind and many times join if a program is joining multiple groups... Didn't read all the multicast rfc files but I guess linux is not violating them. It's late here and I guess I don't explain well so if you have any questions, don't hesitate to ask me. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira