AT wrote:

NIC driver problem? hmm... the NIC brand was BMC.

broadcom?




try this setup: use ETH0 for your typical
LAN connection and as well as your heartbeat
medium. while use ETH1 as your dedicated disk
replication (and OpenMosix Cluster) network.
i define ETH) as my LAN connection and ETH1 as my dedicated heartbeat,
failover doesnt takes place - nothing happens if i stop the heartbeat
service on my primary cluster or if I yank the cable of my primary cluster.

of course, nothing will happen if remove the LAN cable connected on eth0. IP take over will only occur you'll remove the LAN cable on eth1. you also have to add the IPFAIL parameter on your HA.CF so that HA can detect which of the nodes is experiencing a network failure especially if you're going to use a non-ip heatbeat or using a different NIC aside from eth0.




-- Philippine Linux Users' Group (PLUG) Mailing List [EMAIL PROTECTED] (#PLUG @ irc.free.net.ph) Official Website: http://plug.linux.org.ph Searchable Archives: http://marc.free.net.ph . To leave, go to http://lists.q-linux.com/mailman/listinfo/plug . Are you a Linux newbie? To join the newbie list, go to http://lists.q-linux.com/mailman/listinfo/ph-linux-newbie

Reply via email to