API Authentication¶

The following document explains how to authenticate with Run:ai APIs.

Run:ai APIs are accessed using bearer tokens. A token can be obtained in several ways:

• When logging into the Run:ai user interface, you enter an email and password (or authenticated via single sign-on) which are used to obtain a token.
• When using the Run:ai command-line, you use a Kubernetes profile and are authenticated by pre-running runai login (or oc login with OpenShift). The command attachs a token to the profile and allows you access to Run:ai functionality.
• When using Run:ai APIs, you need to create an Application through the Run:ai user interface. The Application is created with specific roles and contains a secret. Using the secret you can obtain a token and use it within subsequent API calls.

Create a Client Application¶

• Open the Run:ai Run:ai User Interface.
• Go to Settings | Application and create a new Application.
• Set the required roles:
  • Select Researcher to manipulate Jobs using the Cluster API. To provide access to a specific project, you will also need to go to Application | Projects and provide the Application with access to specific projects.
  • Select Editor to manipulate Projects and Departments using the Administrator REST API.
  • Select Administrator to manipulate Users, Tenant Settings and Clusters using the Administrator REST API.
• Copy the <APPLICATION-NAME> and <CLIENT-SECRET> to be used below
• Go to Settings | General, under Researcher Authentication copy <REALM>.

Request an API Token¶

Use the above parameters to get a temporary token to access Run:ai as follows.

Example command to get an API token¶

Replace <COMPANY-URL> below with app.run.ai for SaaS installations (not <company>.run.ai) or the Run:ai user interface URL for Self-hosted installations.

curl -X POST 'https://<COMPANY-URL>/auth/realms/<REALM>/protocol/openid-connect/token' \
--header 'Content-Type: application/x-www-form-urlencoded' \
--data-urlencode 'grant_type=client_credentials' \
--data-urlencode 'scope=openid' \
--data-urlencode 'response_type=id_token' \
--data-urlencode 'client_id=<APPLICATION-NAME>' \
--data-urlencode 'client_secret=<CLIENT-SECRET>'

import http.client

conn = http.client.HTTPSConnection("")
payload = "grant_type=client_credentials&client_id=<APPLICATION-NAME>&client_secret=<CLIENT_SECRET>"
headers = { 'content-type': "application/x-www-form-urlencoded" }
conn.request("POST", "/<COMPANY-URL>/auth/realms/<REALM>/protocol/openid-connect/token", payload, headers)

res = conn.getresponse()
data = res.read()
print(data.decode("utf-8"))

Response¶

The API response will look as follows:

{
  "access_token": "...", // (1)
  "expires_in": 36000,
   ....
  "token_type": "bearer"
}

1. Use the access_token as the Bearer token below.

To call APIs, the application must pass the retrieved access_token as a Bearer token in the Authorization header of your HTTP request.

• To retrieve and manipulate Workloads, use the Cluster API. Researcher API works at the cluster level and you will have different endpoints for different clusters.
• To retrieve and manipulate other metadata objects, use the Administrator REST API. Administrator API works at the control-plane (backend) level and you have a single endpoint for all clusters.