Its a system administrator duty to be sure the ESX servers health and performance are OK.
I believe the most important things to monitor are:
VOLUMES (storage) check_vfs
Hardware (Fan,Temp,Arrays,Memory Dimms,Disks etc…)
I have tested all checks with ESX 4, HP or Dell Hardware.
For the hardware checks to work you will need to install
Dell Open Manager or HP Insight Manager
I have other scripts that monitor number of VMs and other details but
those are considered performance and not critical, anyways if someone needs it
let me know.
Just browse the link, I also added description for each one of them.
They can be called by using check_by_ssh nrpe, snmp.