Most of the Cloud Engagement we have, We struggle to monitor the cloud management stack. With vROps we have been monitoring the Effect of Running your cloud on your Infrastructure but not the Cloud Services itself (Well, this is what the vROPs Should Ideally be used for).
For One of the Customer the Requirement was that the customer wanted to Monitor the Cloud Components and Services in one Dashboard and alerting based on the same.
The initial Trial and Error :-
- we initially started by installing EPOPs agents in the windows components and the vRA machine. We were able to monitor the windows services like VCAC Manager Service, Proxy Service etc.
In the VRA OVA there was an problem, The Only 2 services we were able to monitor were Rabbit MQ service and Postgress service.
The Reason :- All other Services we see on the VAMI page are wrapper services and internally call some other service. (Why should Life be easy!!!)
- Then the second option was to try and monitor those services via Log Insight, So, we now deploy VRA Content Pack which says that it will Monitor all the VRA services.
- Deploy LI agents on all the machines, But somehow the content pack does not find the logs that are needed. So we are back to square one.
The Solution :-
After a lot of RnD and searching the Vault and Google (not by me), We find a Management Pack for SDDC Health, This Management pack claims to monitor all the management components in your environment
It provides health of the following
- VMware vCenter health
- VMware vRealize Operations Manager health
- VMware vRealize Automation health
- VMware vSAN health
- VMware vRealize LogInsight
- VMware vRealize Business for Cloud
- VMware Site Recovery Manager
- VMware NSX for vSphere
Seems like exactly what we need, On a Friday afternoon, So we install EPOPs agents on all the management components and then the Management Pack in vROPs,
Installation is As simple as Next à Next à Finish. And to our surprise the Management Pack does not need any Configuration, It Picks up all the Management components itself, Apart from NSX. As NSX does not have EPOPs Agents installed in it (And I hate NSX, and People who work on it, 😛 )… But for VRA Components it covers all the major services.As soon as the Management Pack is installed it starts Collecting provided you have installed EPOPs agents in all the machine.
There are 2 out of the Box Dashboards, that start populating immediately, the Dashboards are simple but do the job.
And as promised all the services are monitored as well
All the collected Metrics are Grouped in various Object types which makes it easy to Navigate (Please Kill me for writing that… Try creating Alert any you will know how easy it is)
Now the Next step was to Create Alerts for all the Service, Which was again a simple task,
Create Symptoms: -As I said it is simple task as the Objects will not be available under SDDC Health but under EPOPS Adapter (If you still don’t get it here is a board that says “sarcasm”…)
Create Alert :- One you have created all the Symptoms that is Around 14, You Put them all in one single Alert.
And Bang… You are good for a good Friday night party…