Nutanix Enterprise AI Manual: API Client Keys, Audit Events, and Infrastructure Health
API client key lifecycle (creation, activation, deactivation, expiration), audit event logging and event filtering, cluster-level and node-level infrastructure metrics, and Kubernetes cluster health monitoring.
API CLIENT KEYS
API keys are required to access the endpoints to integrate with the application.
Perform the following actions to manage an API key:
Create an API key
Update an API key
Delete an API key
Activate or deactivate an API key
Viewing API Keys in Nutanix Enterprise AI
The API Keys page displays all the API keys that you created in Nutanix Enterprise AI.
About this task
To view the API keys, follow these steps:
Procedure
- Log in to Nutanix Enterprise AI.
2. From the left navigation pane, select
API Client Keys
.
API Client Keys
The
page opens, displaying a summary of all the API keys that you created. For information on
the attributes of the API keys displayed in the page, see
API Keys Attributes
on page 282.
3. To view only the endpoint keys, click the
Endpoint Keys
tab.
4. To view only the MCP Client keys, click the
MCP Client Keys
tab.
API Keys Attributes
The following table describes the API key attributes that appear on the
API Keys
page.
Table 53: API Keys Attributes - Description
| Property 1 | Field | Description | Values |
|---|---|---|---|
| Key Name | Displays the user-provided name when creating the key. | Key name | |
| Key Value | Displays the API key created by the user. The key is masked. | Key | |
| Endpoints | Displays the name of the endpoint associated with the API key. | Endpoint name | |
| Created By | Displays the name of the user who created the key. | User name | |
| Status | Displays the current status of the API key. | Active, Inactive | |
| Creating an API Key |
Create an API key in Nutanix Enterprise AI to allow an application to access an inference endpoint.
About this task
To create an API key, follow these steps:
Procedure
- Log in to Nutanix Enterprise AI.
API Client Keys
2. From the left navigation pane, select
.
API Client Keys
The
page opens, displaying a summary of all the API keys that you created. For information on
the attributes of the API keys displayed in the page, see
API Keys Attributes
on page 282.
3. Click
Create a New Key
.
Create API Key
The
dialog box opens.
4. In the
Key Name
field, enter a name for the API key.
Nutanix recommends that you provide a name that is meaningful and identifiable to you.
5. To assign the key to an endpoint, from the
Unified Endpoints
or the
Local Endpoints
dropdown menu,
select the endpoint.
6. Click
Create
.
API Key Details
The system creates the API key and opens the
dialog box that displays the details of the API
key.
7. Click
Copy Key
.
- Store the key securely.
You cannot view the key again after you close the
API Key Details
dialog box. If you lose the key, you must
generate a new one.
Activating or Deactivating an API Key
Activate or deactivate an API key that you created in Nutanix Enterprise AI.
About this task
By default, an API key is activated when you create it.
To activate or deactivate an API key, follow these steps:
Procedure
- Log in to Nutanix Enterprise AI.
API Client Keys
2. From the left navigation pane, select
.
API Client Keys
The
page opens, displaying a summary of all the API keys that you created. For information on
the attributes of the API keys displayed in the page, see
API Keys Attributes
on page 282.
3. Perform one of the following:
»
Actions
To activate an inactive API key, select the checkbox associated with the key, and from the
dropdown
menu, click
Activate
.
»
To deactivate an active API key, select the checkbox associated with the key, and from the
Actions
dropdown
Deactivate
menu, click
.
The system prompts you to confirm the action.
4. Click
Confirm
.
The system displays the new status of the key.
Updating an API Key
Update an existing API key that you created in Nutanix Enterprise AI.
About this task
To update an API key, follow these steps:
You can perform the following API key configuration:
Assign the same API key to multiple endpoints.
Assign the same API key to multiple unified endpoints.
Remove an endpoint to which the API key is assigned.
Remove a unified endpoint to which the API key is assigned.
Procedure
- Log in to Nutanix Enterprise AI.
2. From the left navigation pane, select
API Client Keys
.
API Client Keys
The
page opens, displaying a summary of all the API keys that you created. For information on
the attributes of the API keys displayed in the page, see
API Keys Attributes
on page 282.
3. Select the checkbox associated with a key, and from the
Actions
dropdown menu, click
Update
.
Update API Key
The
dialog box opens.
4. Perform one of the following actions:
»
Endpoints
To assign the API key to an endpoint, from the
dropdown menu, select the checkbox associated
with the endpoint.
»
Unified Endpoints
To assign the API key to a unified endpoint, from the
dropdown menu, select the
checkbox associated with the unified endpoint.
»
To unassign the API key from an endpoint, from the
Endpoints
dropdown menu, clear the checkbox
associated with the endpoint.
»
To unassign the API key from a unified endpoint, from the
Unified Endpoints
dropdown menu, clear the
checkbox associated with the unified endpoint.
You cannot change the key name or the key.
5. Click
Update
.
Deleting an API Key
Delete an existing API key that you created in Nutanix Enterprise AI.
About this task
To delete an API key, follow these steps:
Procedure
- Log in to Nutanix Enterprise AI.
API Client Keys
2. From the left navigation pane, select
.
API Client Keys
The
page opens, displaying a summary of all the API keys that you created. For information on
the attributes of the API keys displayed in the page, see
API Keys Attributes
on page 282.
3. Select the checkbox associated with an API key, and from the
Actions
dropdown menu, click
Delete
.
The system prompts you to confirm the delete action.
4. In the field provided, type
delete
and click
Delete API Key
.
API Keys
The API key is deleted from Nutanix Enterprise AI and is no longer displayed on the
page.
VIEWING AUDIT EVENTS IN NUTANIX ENTERPRISE AI
You can view a list of all the event messages audited and captured as part of the audit logs on the Audit Events page of Nutanix Enterprise AI. By default, no retention period exists for the event messages.
About this task
Event messages describe actions such as importing a model, resetting the password, a user logging in and logging out, updating or deleting an endpoint, or creating an endpoint, API key, or user. Unlike alerts, event messages are simply informational without the need to acknowledge or resolve.
To view the event messages, follow these steps:
Procedure
- Log in to Nutanix Enterprise AI.
2. From the left navigation pane, select
Audit Events
.
The
Audit Events
page opens, displaying a list of all the event messages that are audited and captured as part of
Audit Events Summary
the audit logs. For information on the fields that appear on the page, see
on page 286.
Audit Events Summary
View the list of event messages in Nutanix Enterprise AI.
Audit Events
The following table describes the fields that appear on the
page.
Table 54: Audit Events Summary - Field Description
| Field | Description | Property 3 | Values |
|---|---|---|---|
| Description | Displays the name of the event. | Event name | |
| Action | Displays the type of operation that took place. The possible operation types depend on the entity type. | Create, Delete, Log In, Log Out, Reset Password, Update, Expire, Pulse Upload Event | |
| Entity Type | Displays the entity type, such as an endpoint, cluster, model, and so on. | API Key, Cluster Config, Endpoint, Hugging Face Token, Model, NVIDIA NGC API Key, License, User | |
| User | Displays the name of the user who performed the event. | User name | |
| Time | Displays the date and time when the event occurred. | Date and time | |
| Audit Events Filters | |||
| Audit Events |
The following table describes the filters that are available on the
page. For information on how to
filter the list of event messages, see
Filtering Audit Events
on page 287.
Table 55: Audit Events Filters - Field Description
| Property 1 | Filter | Description | Property 4 | Property 5 | Values |
|---|---|---|---|---|---|
| Entity Type | Select the checkboxes of one or more entities to filter for actions on those entity types. | API Key, Cluster Config, Endpoint, Hugging Face Token, Model, NVIDIA NGC API Key, License, Support Telemetry, User | |||
| Action | Select the checkboxes of one or more actions to filter for those actions on an entity. | Create, Delete, Log In, Log Out, Reset Password, Update, Expire, Pulse Upload Event | |||
| User | Enter a user name and press Enter to filter for actions requested by that user. | User name | |||
| Click | + Add Rule | to create a new | |||
| rule. | |||||
| Filtering Audit Events | |||||
| Filter the list of event messages in Nutanix Enterprise AI. | |||||
| About this task | |||||
| To filter the event messages, follow these steps: |
Procedure
- Log in to Nutanix Enterprise AI.
2. From the left navigation pane, select
Audit Events
.
The
Audit Events
page opens, displaying a list of all the event messages that are audited and captured as part of
the audit logs.
3. Click
Modify Filters
.
The
Filters
pane opens, displaying the available filters. For information on the available filters, see
Table 55:
| Audit Events Filters - Field Description | on page 287. |
|---|
- To view the event messages based on a filter, select the checkbox associated with the filter.
You can apply multiple filters together. For example, if you want to filter for endpoint creation, you can select the
Endpoint
checkbox in
ENTITY TYPE
and then select the
Create
checkbox in
Action
.
5. To export the filtered event messages, follow these steps:
a. Click
Export
.
b. Select
All entities
or
Only Filtered Entities
.
c. Click
Export
.
VIEWING INFRASTRUCTURE USAGE METRICS
View the infrastructure usage metrics of the Kubernetes cluster that hosts Nutanix Enterprise AI and of the individual nodes in the cluster.
About this task
To view the infrastructure usage metrics, follow these steps:
Procedure
- Log in to Nutanix Enterprise AI.
2. From the left navigation pane, do one of the following:
Dashboard
Select
.
The system displays the dashboard with the widgets.
From the
Infrastructure Summary
widget, click
View Details
.
Infrastructure
Usage
Select
.
The system displays the
Usage
tab in the
Infrastructure
page.
Cluster Usage Metrics
For information on the cluster usage metrics, see
on page 288 and for node usage
statistics, see
Node Usage Metrics
on page 289.
Cluster Usage Metrics
Reference for the cluster usage and health metrics on the Infrastructure page when you select Usage and then Clusters .
Infrastructure
Usage
The following table describes the metrics that appear in the
page when you select the
tab and
then select
Clusters
from the drop-down menu on the left.
The graphs display No Data when there is no data for a specific interval.
The graph is displayed for a specified
you select from the dropdown menu on the right of the widget.
interval
Last 15 minutes
Last 1 hour
Last 24 hours
The
involves
,
, and
.
interval
Table 56: Cluster Usage and Health - Metrics Description
| Property 1 | Metrics | Description | Property 4 | Property 5 | Property 6 |
|---|---|---|---|---|---|
| Memory Usage | Displays the graph of memory capacity currently being used by the Kubernetes cluster that hosts Nutanix Enterprise AI. | ||||
| CPU Usage | Displays the graph of CPU capacity currently being used by the Kubernetes cluster that Nutanix Enterprise AI. | ||||
| Metrics | Description | ||||
| Nodes | Displays a summary of all the nodes in the cluster that hosts Nutanix Enterprise AI, along with their names, number of GPUs, CPU capacity, and memory capacity currently in use. | ||||
| Node Usage Metrics | |||||
| The following table describes the metrics that appear in the | Infrastructure | page when you select the | Usage | tab and | |
| then select a node from the drop-down menu on the left. |
The graphs display No Data when there is no data for a specific interval.
The graph is displayed for a specified
you select from the dropdown menu on the right of the widget.
interval
The
involves
Last 15 minutes
,
Last 1 hour
, and
Last 24 hours
.
interval
Table 57: Infrastructure Usage and Health - Node Metrics Description
| Property 1 | Widget | Description |
|---|---|---|
| Memory Usage | Displays the graph of memory capacity used by the node. | |
| CPU Usage | Displays the graph of CPU capacity used by the node. | |
| GPU Utilization | Displays the graph of GPU capacity used by the node. | |
| GPU Memory Usage | Displays the percentage of GPU memory capacity used by the node. | |
| Viewing Kubernetes Cluster Resources and Health Metrics |
View the resources available in the Kubernetes cluster that hosts Nutanix Enterprise AI.
About this task
To view the Kubernetes cluster resources, follow these steps:
Procedure
- Log in to Nutanix Enterprise AI.
2. From the left navigation pane, select
Infrastructure
.
The
Overview
tab in the
Infrastructure
page opens, displaying a summary of the resources available in the
Kubernetes Cluster Summary
Kubernetes cluster. For information on the fields that appear on the page, see
on
page 289.
For information on the health and infrastructure usage statistics of the cluster and the nodes, see
Viewing
Infrastructure Usage Metrics
on page 288.
Kubernetes Cluster Summary
Kubernetes Cluster
The following table describes the fields that appear on the
page.
Table 58: Kubernetes Cluster Summary - Field Description
| Field | Description |
|---|---|
| Name |
Displays the name of the node in the cluster
Displays the health of the node.
Green dot indicates node is online
Grey dot indicates node is offline.
Version
Displays the Kubernetes version running on the node pool.
vCPU
Displays the number of vCPUs assigned to the node pool.
Memory
Displays the memory available to the node.
Disk
Displays the storage space available to the node.
Accelerators
Displays the name and number of GPUs assigned to this node when you hover over the hover info icon.
GPU Memory
Displays the total GPU memory available on the node.
The number of GPUs displayed might be incorrect if the GPU operator pods in your Kubernetes cluster are unhealthy. To fix these unhealthy pods, contact NVIDIA support.
Last updated Oct 08, 2026