vRealize Operations Manager – vROPS 6.0 – New Deployment
In this post we will go through the steps of deploying vROPS 6.0 and configuring basics to allow vROPS to start collecting data from vCenter instance.
Some notes about vROPs deployments in general:
- The resources that is needed for VROPS is dependent on multiple factors, such as number of management packs, objects (hosts, VMs, etc)
- During installation, each node can be Extra Small/Small/Medium/Large in size.
- Every node must be same size.
- Each node must be registered and resolved by DNS (forward/reverse)!!!
- No requirement for IP Pools
Quick overview of installation steps:
Installation of Master Node > Configuration of Master Node > Optional – Installation of Data Nodes and Remote collectors > Optional – Enabling HA –> Login into UI >> Configuration for new installation or upgrade scenarios
- Single location with one or more vCenters: Scale out, multi node vROPs, scaling out as necessary in the same site
- Multiple location/sites and one or more vCenters at each site
- At each location, vROPS instanca can be deployed. In this case, each user log into their respective vROPS instance at their site.
- VROPS cluster at main Datacenter and Remote Collectors at remote sites.
- Each user logs into VROPS UI at the main site
- If the network between sites is not at the speed of LAN and you want to have single vROPS instance, remote collectors must be used to collect data from other sites.
- Two main locations with one or more vCenters: Seperate VROPS clusters at each site can be deoployed. Each cluster/instance owns its data and do not share data between deployments.
For sizing, please refer to the article which is being maintained by VMware: vRealize Operations Manager 6.0 Sizing Guidelines (2093783)
As of Feb 27th, the following is the recommended sizing for vROPS. Please check out the KB for up to data…
Deploy OVF Template
Installation is very easy. We will be deploying ovf as usual via vCenter. After pointing the ovf file, in the second step, accept “extra configuration” to allow numa.vcpu.preferHT to be enabled.
Select deployment size (in this case small < 2000 VMs). This will deploy appliance with 4vCPU and 16GM memory.
For the storage part, it depends on your environment but the best practice is to have thick eager zeroed since the workload will be write intensive.
For Network, please use static IP if you are not deploying the appliance just to play :=)
Configuring the vRealize Operations Manager Cluster
After you deploy the appliance, through web browser, open DNS name of the deployed system.
- Express: Quick and easiest. Best for 1 node vROPS clusters.
- New Installation: Advanced Configuration options that is not available with express options
We will be going with new installation. The first node will be Master.
As a best practice, provide external NTP source for master node. All other nodes will sync time from Master node:
Now you can start Cluster, this will take some time though. In addition, as shown in the initial workflow, you can now optionally add additional nodes to this instance as either Data node or Remote Collector.
Add other nodes:
Open the URL for the additional node and select Expand an Existing Installation:
For shared password, you can create one if someone else is deploying additional node and you don’t want to give the admin password. For this, in the Master Node UI, click generate password icon…
We will be configuring vCenter Solution to collect data. There is two adapters for vCenter.
One is vCenter Adapter to collect data and the second one is python Actions Adapter to apply/Remediate resolution actions via vROPS. Add Adapter for each and click save button before proceeding further.
For monitoring goals in the next section, I will be going with the following policy: All Objects, All Alerts, Dont Overcommit, Include Network and Storage IO..
Now, once you click finish you should be able to see adapters in “Data Receiving” State.
An Voila! Of course, it needs some time to collect data and analyze.