Why no CIM in regular ESX 3.5 U2?

[Svedja]

Hello.

I'm just wondering why CIM isn't implemented in regular ESX 3.5? Or have I missed something in our setup?

Space shouldn't be any problem (as it is larger than ESX3.5i anyway) and the software is there in ESX3.5i already.

As it is now, if you have mixed ESX3.5i and "regular" ESX3.5 you must have two different ways of monitoring hardware, CIM and whatever you hardware vendor has as monitoring programs.

I would like to minimize the diversity/complexity, especialy if we were to move to some kind of ESX3i in the future.

Second reason would be if I could monitor the hardware thru VC if we implement DPM power managment.

I'm still pondering a way of distinguising "sleeping" server from just plain old "dead" server when in powersaving mode.

As VC "knows" the corret state, the monitoring program could rely on the information and we could implement DPM without getting a lot of false failures.

[kjb007]

While a CIM server is included (Pegasus), I haven't seen any references to anyone using it. It provides an SDK interface, which I assume is the reason that not many are using it. I could be mistaken here, but the only material I've found for it is references to SDK and C#/C++. That's about all I have on that.

-KjB

[Svedja]

I've noticed pegasus (even had problems with it during update 1, as many others) but it doesn't seem used, not even by VC.

As I said I would love to have it report health back to ESX & VC, so you could monitor the health in VC client as with ESX3i.

[kjb007]

I agree with you there. While the capability exists, there's no practical use of it that I've seen so far. It may be a future plan.

-KjB

[matuscak · Accepted solution]

Assuming you're talking about the "Health Status" in the configuration tab, if you disconnect and then reconnect the host in VC, the option magically appears. VC must have some data structure cached someplace so it doesnt show up after the update.

[kjb007]

Nice tip here. I didn't know you had to disconnect/reconnect for those to show up. Strange that a reboot was enough.

-KjB

[Svedja]

Nice, thats it. Exactly what I was looking for.

Now, I just have to find a way to use the information for monitoring from our Nagios.

Actualy, I missed it at first after reconnecting, as I expected a new tab (not just selection) when selecting the host.

Thanks for the tip, and I hope that Vmware will fix that problem with health status not appearing.