VMware’s vSphere HA can be found in nearly every data center. In short, HA provides a means for virtual machines to restart on functioning hosts in the event that their current host fails. It’s cool stuff that will save the day in the event of a hardware failure (on the host side, at least). But it also has another role: determining whether an isolation event exists.
Isolation events occur when hosts within an HA cluster lose connectivity to either the network or other hosts in the cluster. We’ve got three options to select from, in order to tell the host how to deal with its VMs: Leave Powered On, Shutdown, and Power Off.
By default, vSphere HA will use the default gateway for its management network as the default isolation address. However, you are able to specify additional isolation addresses for the host to ping to determine whether an isolation response should be triggered. A common and practical design is to include the IP address(es) of your IP-based storage as additional isolation addresses. The goal is to provide enough additional isolation addresses to all but rule out the likelihood of a false positive (and therefore an unnecessary isolation response).
To set these additional isolation addresses, use the following syntax in the Advanced Options screen for your cluster settings:
das.isolationaddress0
You can configure up to 10 addresses (0-9).
For anyone running a lab on a disconnected network, you’ll probably notice that vSphere HA cannot ping the default isolation address (well, if you’ve got no default gateway, that is). In this case, you’ll want to use an additional advanced option: das.usedefaultisolationaddress. Set this to false, then use the das.isolationaddress0 option to specify a valid address on your lab network.
Hope this helps!
In case of multiple isolation addresses, when the HA agent could reach the one of the addresses in between will it continue to check for next address or it stops there.
Thanks for sharing this knowledge..