[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_12362462 ] Bela Ban commented on JGRP-515: ------------------------------- [Regarding my comment on May 17 11:02am] We currently use the wildcard address for receiving IP multicasting packets, so the socket bound to 45566 is 0.0.0.0:45566. This is the same as receive_on_all_interfaces, as traffic will be received on *any* NIC. MulticastSocket.set(Network)Interface() is only used for *sending* packets. So, at one point, we need to decide whether to discard receive_on_all_interfaces, or whether to really bind to the interface defined by bind_addr. The down side of the latter is that we might have more cases where members don't find each other because they bind to different NICs. In any case: even if we receive IP multicast traffic on *all* interfaces, the different multicast (class D) addresses should discard traffic sent to 228.0.0.2:50000 but we are listening on 228.0.0.1:50000 ! This is a bug, as described below: Yes, Aleksander Kostadinov wrote a C test that shows it; you can see it on https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=231899. Just to clarify; this problem does exist on Linux (not just Red Hat; we saw it on SuSE as well). We don't see it on Windows XP and Solaris. If you want to test your OS you can have a look at the C test attached to the above bugzilla, or use the Java-based test attached to http://wiki.jboss.org/wiki/Wiki.jsp?page=PromiscuousTraffic. -- 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