Hello,
We're having some serious issues with our HA (Active/passive) on VMware 5.5.
It started out with high cpu and a constant utilization on the HA sync interface (200Mb/s). When we shutdown node 2, it the cpu went down to normal. We opened a ticket and Sophos instructed us to re-create the cluster. So we did and it actually solved the issue, after some hassle with two of the network interfaces that had to be switched around in vmware on node 2.
This was until we tried a failover yesterday, which resulted in loss of 2 of our 4 networks (HA sync and LAN stayed intacted, DMZ and WAN was unaccessible). Once we were able to fail back to node 1 and got all the networks accessible again, we also went back to square one with the high cpu and ha sync utilization! I'm suspecting that our WAN and DMZ is being switched around somehow on node 2, not sure if this occurs after vmotion or something else that triggers this. The networks are not in the exact same order on the VM's, but they are in the order that we got node 2 to work in the first place. The virtual_mac has been set to 0.
Anyone got any ideas? Are some of the eth interfaces beeing switched due to change of esxi's mac change?
This thread was automatically locked due to age.