The number of heartbeat datastores for host is 1, which is less than required: 2
The number of heartbeat datastores for host is 0, which is less than required: 2
I ran into these errors recently after configuring High Availability on a brand new VMware ESXi 5.1 cluster. I had used SanDisk cards for my ESXi operating system and planned to use a LeftHand Virtual SAN Appliance with only one LUN so I was only going to ever have one shared datastore. I knew I would have to squelch this error right away.
This error occurs because in VMware vSphere 5.0 they have made new advancements in high availability called datastore heart beating.
In earlier versions of VMware vSphere such as 4.1, if a host became unreachable on the management network it would be considered down and all the high availability mechanisms would go into full swing. Even though the host is unreachable on its management network it was still possible the host was still running and serving out virtual machines. This condition is referred to as host isolation. VSphere 4.1 misidentifying this outage would cause unnecessary stress by moving the virtual machines to a new host.
In VSphere 5.0, datastore heart beating was the answer to this problem. All hosts write a small file to the shared datastore. If the host can’t be reached on the management network, vSphere checks to see if it is updating its lock file on the shared datastore. This way it can differentiate between a failed and an isolated host.
To squelch this error, however, if you only have one shared datastore, check out this VMware KB article.
Leave a Reply