I’ve seen this one on the VMTN forum a couple of times. When the Health Status isn’t showing you could do the following to fix it:
- Restart VirtualCenter service on the VC Server
- Restart mgmt-vmware service on the hosts that are affected (service mgmt-vmware restart)
- Restart vmware-vpxa on the hosts that are affected (service vmware-vpxa restart)
If the above did not fix the issue:
- Disconnect the affected Host from Inventory on VC
- Reconnect the affected Host from Inventory on VC
And if that doesn’t work this is also a possible solution:
- Restart the Pegasus service (service pegasus restart)
When this happens for me I go to the top-most level of alarms and adjust one of the two preset VM alarm Definitions in the Yellow trigger by one and then the grey health marks change to where they should be.
Thanks for the great information! Saved me some searching today!
http://www.vmwareinfo.com/2008/08/missing-health-status.html
Disconnecting and reconnecting fixed it for me.
We had just updated VC and ESX to 2.5U2 and 3.5U2 respectively, so all services had just been restarted.
The hosts may go through and inventory process as part of their being connected that creates the health objects.
This issue is fixed via adding the Alarm to heartbeat VM.
And allso combined with an alarm to monitor a VMs State will help determine the satyus of the VM, not the Host..
If you are running HP Proliant or blade hardware it is also worth checking your BIOS revision. We had a number of BL4xx blades which would not show the health status. Updating the BIOS solved this.
Disconnecting / reconnecting only seems to partially fix the issue for me. Only 3 of the hardware monitors are shown. Reinstalling the cim/pegasus rpms solved the problem.
This is such a great resource that you are providing and you give it away for free. I love seeing websites that understand the value of providing a quality resource for free. It?s the old what goes around comes around routine.
FYI, another item to check: When we access vCenter via vSphere Client, the “Health Status” option is missing for every host. But when we access each individual host directly via vSphere Client, “Health Status” shows up fine. So in our case at least, it seems to be an issue with vCenter, not with the hosts.
This issue has remained unchanged while hosts have been added, power-cycled, rebooted, and upgraded. Since we’re planning other health monitoring, we haven’t done any troubleshooting on this, or even restarted vCenter, so no results to share on that.
Our environment: vCenter 4.0, vSphere Client 4.0, a mix of ESX and ESXi 4.0, various HP Proliant servers.
Check in VCenter, Plug-ins, Manage Plug-ins menu and if you see an error for the “VCenter Hardware Status” plugin then you probably cannot reach VCenter on port 8443.
Open your firewall to allow connections to VCenter on TCP 8443 and then you should see the “Hardware status” tab on each host.