ESX Hardware Health on Dell PowerEdge servers

[JayScovill]

I have three Dell R600 servers running ESX. nWorks is reporting the hardware state of these machines as healthy. However, one of redundant power supplies on each of the machines aren't plugged in and this is reported through the vCenter client on each host.

Why is the health state through SCOM reporting the power state as healthy?

Thanks

[Alec King]

Hi Jay,

There is one possibility - if that power supply has ALWAYS been unplugged, then it may be that you had a timing issue when you first deployed the nworks MP.
If a hardware sensor is already red when nworks first detects it, then the change-to-red event can be missed by SCOM - because the first full hardware discovery has not completed yet in the MP.

Note - this can only happen the very first time nworks is installed, and only if a sensor is already red. Once first hardware discovery has completed, all state-change events for hardware will hit the corresponding sensor in our SCOM topology.

There's one way to check this -
Stop the nworks Collector service.
Delete the file StateTbl.xml (default location Program Files\nworks\VIC\Collector\Data)
Start the nworks Collector service again.

This file is the database of all current state objects (such as hardware sensors). Deleting the file will force the Collector to re-post an event showing the status of every sensor.
That should trigger the monitor and set your power supply to red.

Let me know if that fixes it!

Cheers,