---
title: "Nutanix data source plugin for Grafana"
description: "Health, capacity and performance of Nutanix clusters, hosts, VMs and storage containers from Prism Central. See install steps and the changelog."
---

> This is the Markdown version of https://grafana.com/grafana/plugins/integrationplumbers-nutanix-datasource/

# Nutanix

- Developer: integrationplumbers
- Version: 1.0.0
- [Installation](https://grafana.com/grafana/plugins/integrationplumbers-nutanix-datasource/installation.md)
- [Change log](https://grafana.com/grafana/plugins/integrationplumbers-nutanix-datasource/changelog.md)

![Cluster Overview](https://grafana.com/api/plugins/integrationplumbers-nutanix-datasource/versions/1.0.0/images/img/screenshots/cluster-overview.png)

![Host Performance](https://grafana.com/api/plugins/integrationplumbers-nutanix-datasource/versions/1.0.0/images/img/screenshots/host-performance.png)

![VM Performance](https://grafana.com/api/plugins/integrationplumbers-nutanix-datasource/versions/1.0.0/images/img/screenshots/vm-performance.png)

![Storage and Capacity](https://grafana.com/api/plugins/integrationplumbers-nutanix-datasource/versions/1.0.0/images/img/screenshots/storage-and-capacity.png)

![Alerts and Events](https://grafana.com/api/plugins/integrationplumbers-nutanix-datasource/versions/1.0.0/images/img/screenshots/alerts-and-events.png)

Dashboards for the health, capacity and performance of Nutanix clusters, hosts, VMs and storage containers, with their alerts and events. The plugin reads everything from Prism Central's API. Nothing is installed on the clusters. The plugin does not connect to Prism Element.

## Requirements

- Grafana 12.4 or later.
- A Prism Central that serves the Nutanix v4.0 API or the v3 API, reachable from Grafana.
- A Prism Central account with the Prism Viewer role, described below.

## Getting started

1. **Create a credential.** Use an API key on a Prism Central that serves v4.0. Use a username and password on one that serves only v3. The account needs the Prism Viewer role and no other. Credentials, below, has the steps.

2. **Add the data source.** In Grafana, go to **Connections**, then **Data sources**, and add **Nutanix**. Enter the Prism Central address, usually `https://<prism-central>:9440`. Under **Authentication**, choose **API key** or **Username and password** and enter the credential. **Save & test** reports the number of clusters and the Prism Central version.

3. **Import the dashboards.** On the data source's settings page, open the **Dashboards** tab and select **Import** beside each dashboard. Start with **Nutanix Cluster Overview**.

If a private certificate authority signed the certificate at the Prism Central URL, paste that authority's certificate as the **CA certificate**. **Skip TLS certificate validation** connects without verifying the certificate.

To try the plugin without a Prism Central, turn on **Demo mode** in the data source settings. It serves sample data for three clusters and sends nothing over the network.

If Grafana cannot reach Prism Central directly, use Grafana's secure SOCKS proxy or [Private Data Source Connect](https://grafana.com/docs/grafana-cloud/connect-externally-hosted/private-data-source-connect/). The proxy or agent needs to reach Prism Central only.

## Credentials

Prism Viewer is the narrowest built-in role that covers every request the plugin makes. Use an account created for the plugin.

### API key

Prism Central issues API keys to service accounts only. Service accounts, keys and their role are created through the Nutanix IAM v4.0 API, by a Prism Admin or Super Admin. Nutanix's [API key walkthrough](https://www.nutanix.dev/2025/02/21/nutanix-v4-apis-using-api-key-authentication-part-2/) shows the requests.

1. **Create a service account.** Call `createUser` with `userType` set to `SERVICE_ACCOUNT`.

2. **Create an API key for it.** Call `createUserKey` with a `name` and `keyType` set to `API_KEY`. Prism Central returns the key value once.

3. **Give the service account the Prism Viewer role.** Call `createAuthorizationPolicy` with the Prism Viewer role, the service account as its identity, and every entity in scope. The walkthrough's example assigns Super Admin. Assign Prism Viewer.

Paste the key into the data source's **API key** field.

### Username and password

A Prism Central that serves only v3 cannot issue API keys. Create a Prism Central user for the plugin, give it the Prism Viewer role, and enter its username and password.

## Dashboards

| Dashboard                    | Shows                                                              | Select                             |
| ---------------------------- | ------------------------------------------------------------------ | ---------------------------------- |
| Nutanix Cluster Overview     | What needs attention across every cluster, and how each is running | One cluster, or All                |
| Nutanix Host Performance     | The hosts of one cluster                                           | A cluster, and its hosts           |
| Nutanix VM Performance       | The VMs of one cluster, or of one host                             | A cluster, optionally a host, VMs  |
| Nutanix Storage and Capacity | One storage container's capacity, efficiency and performance       | A cluster, and a storage container |
| Nutanix Alerts and Events    | Open alerts, and events over the time range                        | Every cluster, or one              |

Each dashboard has a **Data source** selector. With one data source per Prism Central, the same five dashboards serve all of them.

Cluster Overview opens with an estate summary of open alerts, cluster fault tolerance and degraded hosts. Below it is one row per selected cluster, with health, capacity, CPU, memory and storage I/O. The rows start collapsed. Host Performance repeats its panels for each selected host and ranks the hosts by CPU utilization. VM Performance covers CPU, memory, disk and network for the selected VMs. Storage and Capacity also lists the cluster's disks.

Alerts show what is open now, whatever the time range. Events and statistics follow it. The four performance dashboards open on the last 6 hours, and Alerts and Events on the last 24 hours.

The dashboards refresh every minute. Refresh should be set at one minute or longer for larger deployments.

Importing a dashboard again from the **Dashboards** tab replaces the copy in Grafana. When customizing one, save a copy and edit that.

## Data source settings

**Maximum entities per query** caps the rows an inventory, alerts, events or raw query returns, the hosts a host stats query reads, and the entries a template variable lists. The default is 1,000. A query that reaches the cap shows the capped rows with a notice saying so.

## Alert rules

The plugin includes the two alert rules below, as a provisioning file at `alerting/nutanix.yaml` in the plugin's directory. On a Grafana you run yourself, copy it into Grafana's `provisioning/alerting` directory, set the environment variable `NUTANIX_DATASOURCE_UID` to the Nutanix data source's uid, and restart Grafana. Grafana Cloud does not read provisioning files. There, create the rules through the [alerting provisioning API](https://grafana.com/docs/grafana/latest/alerting/set-up/provision-alerting-resources/http-api-provisioning/) or [Terraform](https://grafana.com/docs/grafana/latest/alerting/set-up/provision-alerting-resources/terraform-provisioning/).

| Rule                                  | Fires when                                                                                 | Checked every | For | Severity |
| ------------------------------------- | ------------------------------------------------------------------------------------------ | ------------- | --- | -------- |
| Host degraded                         | Prism Central reports a host as degraded. One alert per host                               | 5m            | 15m | warning  |
| Cluster below desired fault tolerance | A cluster tolerates fewer failures than its desired fault tolerance. One alert per cluster | 5m            | 15m | critical |

Cluster below desired fault tolerance also fires when Prism Central returns no clusters or cannot be queried. Host degraded does not. Both rules carry the labels `severity` and `source=nutanix`.

A Grafana alert rule can query Cluster stats, Host stats, VM stats, Storage container stats and Alert count. Alert count returns one number, such as the open critical alerts.

An Inventory query can back a rule when **Columns** holds exactly one numeric column, with the text columns that name the entity. Each row is then one alert instance. `degraded` on hosts and `faultToleranceDeficit` on clusters are numeric columns for this.

## Template variables

For building your own dashboards. Create each as a **Query** variable on the Nutanix data source.

| Entity type       | Cluster    | Host    |
| ----------------- | ---------- | ------- |
| Cluster           |            |         |
| Host              | `$cluster` |         |
| VM                | `$cluster` | `$host` |
| Storage container | `$cluster` |         |
| Disk              | `$cluster` |         |

Leave **Cluster** or **Host** empty to list across every cluster or host.

## Support

Email <helpdesk@integrationplumbers.io>.

Use of the plugin is subject to the Integration Plumbers [acceptable use policy](https://www.integrationplumbers.io/legal/acceptable-use).
