It depend how you monitor yours ESX servers.
If you monitor them directly :
You need go to Node detail page of the ESX server and click on "List Resources" button on Node detail resource.
When you do that SAM try to poll CIM_Chassis class trough the CIM protocol. If its successful you should see the Hardware Health of Server check box. It should be checked by default. If its not then check it.
If polling of the CIM_Chassis class failed then SAM is unable to poll hardware sensors on this ESX and you will not see the check box there. In that case you should check configuration of your server.
If you monitor them through the vCenter:
On ESX servers polled through the vCenter is Hardware polled automatically. If you do not see it then you should check configuration on the vCenter. Try locate VMWare api UI by typing this into your web browser on SAM server.
https://{IP addressOfThevCenterServer }/mob
This API is not very user friendly but if you are able to locate the ESX server here and it does show some Hardware statuses it should be polled by SAM.
There is another question. Do you poll those ESX server as VMWare?
If you had add this node before you upgraded to SAM 5.0 and the node was polled as VMWare then SAM will try poll hardware through the CIM (or vCenter) (right now there is no way to switch one hardware polling method to another).
If the node is in fact added as SNMP SAM will try use snmp for polling. In that case you will need install Dell OpenManage Agent(Server) on the ESX server because VMWare do not supply information about hardware through the SNMP. But I presume that is not your case?
If you will be unable to found the problem please create support ticket.
So we poll the ESX servers through vcenter. They were set up that way before we upgraded to SAM 5 RC2. We have ESX 4 and 5 servers, neither have the option of adding Hardware Health.
As for browsing to the https://{IP addressOfThevCenterServer }/mob, that works but I wasn't able to find anything resembling any of our servers.
We do not have the OpenManage agents on those servers. But they are also set to be monitored via SNMP.
So since they are being monitored through vcenter and through snmp, it looks like you are saying we will not be able to get hardware stats?
If you're polling your ESX hosts though the vCenter you won't see the hardware option listed when you click "List Resources" as these nodes tend to be ICMP only nodes. This information is automatically collected by SAM when available through the VMWare API. You will however need to ensure you have the vCenter Hardware Status Plug-in enabled on your V-Center for this information to be available though the VMWare API.
We have that plugin enabled on the vcenter server and if I browse to the esx servers there is a Hardware Status tab that contains valid hardware status. We can manually update the status as well. If we get a chance, we will open a ticket.
We are obviously missing something, not sure what it is at this point. It would be nice if Orion had some kind of debug log on this piece (I'm sure there is). Would be really nice if the GUI showed the status of the vsphere poller other than the basic up / down.
To sum up our situation again:
We have a vcenter server with esx 4 and esx 5 hosts. Orion NPM / APM was polling the vcenter server via SNMP and the vSphere API. Orion was polling the ESX hosts via the vcenter server via the API and additionally via SNMP. We then upgraded to SAM 5 RC2 with no change to the methods of polling. At this poinht we do not see hardware status of any of the esx hosts. We were able to add hardware status of the vcenter server.
John
>>You need go to Node detail page of the ESX server and click on "List Resources" button on Node detail resource.
Is this true for all hardware monitoring? So I will have to do this over 500 times?!?!? You guys do not think of large numbers of devices when you design your UI's.
Agreed. Even for the few physical servers we have, it's a bit of a pain.
It's even more painful when you go to enable it but it's not there to select!
It's like real time process monitor and having to drill down to use it. I should be able to right click on the node - ANYWHERE - the node name is displayed.
It's like real time process monitor and having to drill down to use it. I should be able to right click on the node - ANYWHERE - the node name is displayed.
The Real-Time Process Explorer is located in three different resources on the Node Details Page as pictured below.
ALTeReGo - Appreciate you providing help. Would like to suggest that RTPE be available as a right click option. I envision opening up Orion and drilling down to a node in red and just wanting to right click and see the processes causing the issue.
As for my original purpose of this post, is there any more troubleshooting we can do to figure out why our ESX servers do not show hardware status?
As for my original purpose of this post, is there any more troubleshooting we can do to figure out why our ESX servers do not show hardware status?
I would recommend opening a case with support so we can troubleshoot this issue properly. We'll need to look though your diagnostics as well as perform some investigative troubleshooting steps on your vCenter to determine the cause of your issue.
Is this true for all hardware monitoring? So I will have to do this over 500 times?!?!? You guys do not think of large numbers of devices when you design your UI's.
If you run a scheduled Network Sonar discovery of your existing servers, the next time this scheduled discovery runs any servers that support hardware health information SAM will automatically start collecting this information.
If you run a scheduled Network Sonar discovery of your existing servers, the next time this scheduled discovery runs any servers that support hardware health information SAM will automatically start collecting this information.
Doesn't work for me. I have to manually check each one.
Hi we looked into this issue more deeply and we found that there is bug which prevents the ESX(i) to be monitored for hardware information's. We're investigating this now and hope to have a resolution to it fairly quickly.
Following is from the bottom of the first page of the Monitoring Hardware Health chapter, as the manual stands today:
Hardware monitoring is automatically enabled for VMware ESX servers. Monitoring of VMware ESX servers uses the CIM protocol which should be enabled by default after the installation of ESX hosts. If installing Hardware Monitoring Agent software on a VMWare ESX host, hardware information will be detected by SAM and the checkbox to monitor Hardware Health of Serverswill be displayed, even where hardware information might already be collected via the vCenter using the VMWare API. When this option is displayed for an ESX host, checking it will not change the hardware polling method when Poll for VMWare is also selected.
If this information is not helpful in your situation, I will do some research and get more on the matter. Hope this helps.
Real-Time Process Explorer - There is no "right-click" instruction to access this feature that is documented (or that I'm even aware of).
There are three places within SAM that you can access the RTPE: the Node Details page, the Application Details page, and the Component Details page.
Make sure pop-ups are allowed.
Ok, must have misread on the right click piece. Feature Request - We want to right click on a node and launch the RTPE.
We do see that the option on App details and Component Details but not on Node details. Thx.
