I don't know if I'm in the right place here since I don't see a vFabric Hyperic forum and I'm just starting out on Hyperic as well.
I deployed the vFabric Hyperic 5.0 vApp and am finding the documentation either really bad or just too hard to figure out for a beginner. I'm trying to see what I need to do to have Hyperic monitor vCenter 5.1, VMs and ESXi 5.1 hosts assuming it will do all that.
I read not to install the agent on the vCenter server then read you should install it on the vCenter server. Then I read to install the agent on another computer and edit the agent.properties file but the edits look like they are already there. I also read ESXi will report to Hyperic as is but cant find what I need to do to make that happen. It also seems that you can have the vCenter agent running and then have agents on all the VMs for application monitoring.
If anyone can give me the quickie lesson on what I need to do or point me in the right direction I would appreciate it!
You can install the agent on vCenter but install a Hyperic Agent on a platform that can access vCenter Server over the network. To minimize the load on the vCenter platform, do not install the agent on the same platform as vCenter.
Also you can install agents on your VM's, they will monitor your vm.
Iam having the same issue in configuring this out. I am using the HQ (open source) 5.0 and struggling in setting the server to monitor the vmware enviroment, in particular setting up the keystore, etc.
Could you please share the steps of your resolution?
Thank you in advance
That will be a tough one since it was trial and error to get this to work and I went through a ton of emailing back and forth and webex sessions to get it to work and I dont remember the exact thing that got it working.
I installed the agent on a Windows 7 VM since its not recommended to install it on the vCenter server. The tricky part was getting the agent to work and stay running. The agent.properties file has to be just right and you need an SSL certificate for version 5 at least. I didnt set that part up so I cant help you there and it sounds like that may be what you need help on.
Here are some of the notes that I was given to do the setup. Hopefully it makes sense.
Couple key notes - you need to pre-configure the agent via its properties file:
When doing remote monitoring of vSphere, which is recommended, the adding of the vSphere resource is done manually. So no autodiscovery of vSphere/vCenter.
There is a dependency on having a properly configured agent.properties file. (see below)
Also a more recent dependency in newer versions of Hyperic on having the SSL certificates in place - see doc for this one.
1 - Stop the agent
2 - Wipe the agent's /data directory clean
3 - Remove the platform (that the agent sits on) from the Hyperic inventory
...This will clean your slate and allow you to start fresh.
Then proceed to get the SSL cert in place, and configure the agent.properties file as specified below.
Only then, restart the agent. Allow it to report into Hyperic and add a new platform.
Then manually add the vSphere resource as defined below (Tools -> new server)
Port 443 needs to be network routable for a remote agent to connect to it.
To setup the agent on another platform:
## enables unidirectional communications between HQ Agent
## and HQ Server in HQ Enterprise Edition
In the above block, agent.setup.camIP is the IP address of the Hyperic server.
The agent properties must be setup without using the *default* macros. The reason being that the vsphere plugin will check with this file to find the HQ Servers IP, credentials to log in with, along with the IP/port of the box its sitting on. As part of its auto discovery process, the vsphere plugin will make webservice calls through hqapi to the hq server to handle resource creation for the ESX hosts/vms that it discovers.
Edit the agent.properties file to contain the needed info *before* starting the agent.
If things dont work for whatever reason.
After creating the vsphere resource, the vsphere hqu plugin should show up, after configuring the resource with credentials/sdk api, you should see resources pop up into the recently added portlet within a minute.