Hi Folks,
I gone through VMware Availability Guide and understand all VMware cluster nodes exchanges hear beats every second to monitor liveness of other nodes in the cluster. If any node not receiving Heart Beats from other cluster nodes then only it will try ping the isolation address (gateway by default). If it not reach the gateway then it initiate shutdown of VMs. Hope I'm clear with this concept.
Now come to my environment. All ESXi server are running in blade server. When our network goes down, still I'm able to ping other ESXi nodes in cluster from one node. Which means ESXi to ESXi communication is still there as they are in blade. As it can receive heart beats it won't try to reach the gateway or it will not shutdown the virtual machines. But my virtual machines are still getting down by ESXi servers. This is where I'm stuck up.
Please help me to understand whats going on and what I'm missing? Please let me know if you need more clarification.
Thanks in Advance!!!
Hari.
Hi,
May you need change isolation response to Leave Power On.
Check http://www.yellow-bricks.com/vmware-high-availability-deepdiv/
Please, don't forget the awarding points for "helpful" and/or "correct" answers.
Mauro Bonder - Moderator
Hi.. thank you for your response.. We have already configured leave powered on option after we found our VMs shutdown due to network failure.
But my question is as mentioned in availability guide, since all the ESXi server are able to ping together they should not do host isolation shutdown. But here VMs are getting shutdown by ESX eventhough they are able to ping internally regardless of network availability as those are in blade server,
thank you,
Hari
besides I will surely check the link you have shared.
Did you contact support and had them analyze the logs? If it is possible to send a heartbeat than HA will not powerdown any VMs as isolation response would not be triggered and it wouldn't even try to ping the gateway.
Duncan