Hardware Monitor and HP blade server on c7000

Hiyas

Having a problem with Hardware Monitor and some, not all, blades in a HP c7000 chassis. We had a blade that had some bad memory, and Orion reported memory in yellow as "degraded", as I understand it there were also errors and events both in VMware and the HP manager(i am no server guy.. thank god). The server team replaced the affected memory and the errors all cleared up on their side, however hardware monitor in Orion is unchanged and still shows a yellow warning for memory being degraded.

So I turned my attention to the server and sat down with one of our server guys and tried to locate the problem, he showed me their logs and indeed everything on their side appears to be ok. No errors. Also according to the SAM admin guide on page 394 there is some sort of hardware monitoring agent on the HP possibly running on port 2381....

Accessing Hardware Monitoring Agent Software

Each vendor’s hardware monitoring agent software includes a web server that operates on a unique port. Navigating to https://{remote.ip.address}:{Port} successfully will validate that the agent software is installed. Refer to the table below for the list of ports used by each vendor.

Vendor: HP Dell IBM

Port: 2381 1311 423

For example: Navigating to https://10.186.16.100:1311 in your web browser will take you to the Dell OpenManage Server Administrator page on the remotely monitored host.

Actual ports open on the blade IP are 80,427,443,902,5989,8000,8100. I understand that 5989 is CIM.

When I run an NMAP against this blade none of these ports are listening so this software doesn't appear to be running. I mentioned this to our server team and they were no aware of a "hardware monitoring agent" on the blades, but if it isn't running then how is Orion getting its metrics?

SO the problem I am faced with now is as any particular HP blade has a problem it reports that problem to Orion and then that error is forever stuck. We don't seem to have the ability to clear the error or reset it after the condition has been cleared on the other side.

very confused

-e

Find more posts tagged with

cim

i_like_pie

c7000

hardware_monitor

Accepted answers

aLTeReGo

Based on the information you provided it sounds like the managed node is running VMware and not Windows/Linux so the web interface instructions from Administrators Guide are no applicable. For hardware health information on VMware nodes we use CIM or the VMware API. When you open the vSphere client and look at the hardware status for this node what do you see? <Please post a screenshot of the memory category expanded from vSphere> What does SAM show for the same node? <Please post screenshot of SAM Hardware Details resource with memory category expanded>

All comments

aLTeReGo

Bahlkris

Ahhh ok, thanks for pointing me in a direction. So I talked a admin into going into that and it turns out there is a warning under memory in the hardware status. Sorry I don't normally have access to this and trusted someone telling me everything was ok. My hunch here is the warning is due to the memory status being "unknown" per the screen shot.

aLTeReGo

The reason why SAM is reporting memory in a warning state is because that's what VMware is reporting. Why VMware is reporting the memory being in a warning state is no really clear, but it's probably something I'd recommend running down in VMware's forums or by contacting their support. It may be something as simple as clearing the hardware logs through BIOS.

Bahlkris

Right, I don't think we necessarily expect Orion to tell us why the error is occurring, more that it is. My problem in all of this was I was being told there were no errors when in fact there were. At closer inspection over the last 12 hours it appears the firmware on that c7000 is several versions behind.