The objective of this document is to give you the best practices to put in supervision of your Nutanix Cluster in order to realize a service weather representative of its state.
Prerequisites
The use of our Nutanix service models requires the following prerequisites:
- A read-only Nutanix Prism login account used to query the Nutanix API
- SNMPv2 or SNMPv3 configuration information
Nutanix architecture, concepts and components
Your Nutanix cluster tightly integrates processing, storage, networking, and virtualization components.
The Nutanix Virtual Computing Platform is a scalable, converged compute and storage system specifically designed to host and store virtual machines.
All nodes in a Nutanix cluster converge to provide a unified pool of tiered storage and present resources to virtual machines for transparent access. A global data system architecture integrates each new node into the cluster, allowing you to scale the solution to meet your infrastructure needs.
The base unit for the cluster is a Nutanix node. Each node in the cluster runs a standard hypervisor and contains processors, memory, and local storage (SSDs and hard drives).
A Nutanix Controller virtual machine runs on each node, allowing local storage to be pooled from all nodes in the cluster.
Configuration
Setting up supervision of each of the components of the Nutanix architecture
Monitoring the Nutanix Cluster
Monitor the status and performance of your Nutanix Cluster by adding service models to measure its performance:
- Nutanix-Cluster-Status
- Nutanix-Cluster-IOPS
- Nutanix-Cluster-Latency
- Nutanix-Cluster-IOBandwidth
Monitoring the capacity and performance of Nutanix Cluster Storage
Monitor the overall usage of Nutanix Cluster Storage using service models to measure its overall usage:
- Nutanix-Cluster-StorageUsage
A Nutanix Cluster Storage cluster is composed of Containers grouping one or more Storage Pool. In case of an alert on the Nutanix-Cluster-Storage Usage, it will be quickly possible to know the cause by also supervising the Containers and Storage Pools using the following service models:
- Nutanix-Container-Usage
- Nutanix-Storage Pool-Usage
Supervision of the Nutanix blocks that make up the cluster
Ensure the health of your Nutanix Blocks by using the following service templates:
- Nutanix-Disk-Status
Monitoring the Hypervisors of each Nutanix Node
Also make sure that the hypervisors of each Nutanix Node.
When using VMware technology, you can use the following service models to monitor the health of your ESXs:
- VMware-ESX-CPU
- VMware-ESX-Datastore
- VMware-ESX-DiskIO-Read
- VMware-ESX-DiskIO-Write
- VMware-ESX-Hardware
- VMware-ESX-NetUsage
- VMware-ESX-RAM
- VMware-ESX-Runtime_Issues
- VMware-ESX-Runtime_status
- VMware-ESX-Services
- VMware-ESX-Services-WithExclusion
- VMware-ESX-SWAP
Model your Nutanix cluster with the ServiceNav monitoring tool.
Once the supervision is done, you can model your Service Weather.
The ServiceNav Service Weather feature allows you to graphically model your Nutanix Cluster.
Through your weather services that you have configured following our recommendations and best practices available hereIf you have a Nutanix Cluster, you will be able to map the various components of your Nutanix Cluster identically:
The capture below is an example, we find the various elements constituting a Nutanix Cluster including :
As shown in the screenshot below, the interest of such a model is to be able, in case of an alert, to directly access the root cause of a Nutanix Cluster degradation. The screenshot below clearly shows that this degradation is due to a Critical alert in Pool Usage 1 of Container 1
You are able to mobilize the appropriate teams to work on your Nutanix storage and increase your chances of resolving the problem and getting back to normal quickly.
Reproduce your Nutanix service weather through our Dashboard mapping module by following our available best practices here
