VMware – persistent “Host memory status” alarm in vSphere

If you just vMotion your VC/VCUM VM to a different host you won't have any problems, it's how I did my lab boxes in fact. If you don't have vMotion capability then I'd strongly recommend manually moving them to a different host, trying to do it on the same host as the VC is a recipe for disaster as you'll lose your VCUM mid-process.

This may be similar to my other question about specific use-cases of HP SIM at: https://serverfault.com/questions/284004/integrating-hp-systems-insight-manager-into-an-existing-environment

Either way, HP Systems Insight Manager is free and available from:
http://h18013.www1.hp.com/products/servers/management/hpsim/download.html

You can access ILO interfaces detected on your network from one place. HP SIM will also associate the ILO interfaces with the physical servers for you.

The software should probably run on its own Windows (or Linux) server, but for the size of the environment you're describing, it can run in a VM. Here are the minimum hardware specs from the Quickspecs page:

Any HP ProLiant with the following configuration:

Minimum: 1.5-GHz processor with 4GB RAM (for x86) / 6GBRAM (for x64)
1 GB free disk space recommended
Recommended: 2.4-GHz processor with 6GB RAM ( for x86) / 8GB RAM(for x64)

As for monitoring VMs, HP SIM is aware of VMWare ESX and ESXi and will display information about VMs discovered on your host servers. The HP-specific ESXi version will show hardware status and won't require any agents. They've already been integrated into the ESXi build.

**A note on the ESXi builds. I don't believe you'll get full hardware health info from any systems earlier than a G5. You won't be able to get ESXi 4 to run on the G3 servers (not 64-bit capable).