All,
   I know there's probably a simple solution to this, but I don't know what
it is...

I have 2 identical hosts (T6340) built on consecutive days, running S10u7
(141414-10), in the same subnet, connected to the same Cisco switches(with
link aggregation), with Cisco router/firewall.

An application requires multi-cast addressing for it's clustering component,
and it's not working as desired.

To test this I have (most recently) been running "java -cp tangosol.jar
com.tangosol.net.MulticastTest -group 237.152.0.11:6666 -delay 5", although
iperf also produces the same results.

Of the two hosts, each allways sees it's own multicast packets, and one
occasionally (every 15 minutes) sees a packet from the other host.  If snoop
is run on either host, then that host starts receiving packets from the
other within ~90 seconds, and once snoop is terminated the packets stop
being received within ~300 seconds.

It would appear that snoop knows what I need to do/enable/disable/... but I
don't.

Can anybody enlighten me?

Thanks,
Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
<http://mail.opensolaris.org/pipermail/ug-msosug/attachments/20091208/fe08e6a1/attachment.html>

Reply via email to