Azure Red Hat OpenShift Management Client Library for Python

Warnings

• breaking Starting with version 1.5.0, this package exclusively targets the latest Azure API-Version, removing APIs for older versions. Applications requiring a specific, non-latest API-Version should pin to an earlier package version.
  Fix: If your application relies on a specific older API version, pin the `azure-mgmt-redhatopenshift` package to a version prior to 1.5.0. Otherwise, ensure your code is compatible with the latest API version.
• breaking The credential system was completely revamped. Old `azure.common.credentials` or `msrestazure.azure_active_directory` instances are no longer supported. The `credentials` parameter was renamed to `credential`.
  Fix: Migrate your authentication to use classes from the `azure-identity` library (e.g., `DefaultAzureCredential`). Update client instantiation to use the `credential` parameter instead of `credentials`.
• breaking Operations that previously returned `msrest.polling.LROPoller` now return `azure.core.polling.LROPoller` and are prefixed with `begin_` (e.g., `create_or_update` becomes `begin_create_or_update`).
  Fix: Adjust method calls to use the `begin_` prefix for long-running operations and expect `azure.core.polling.LROPoller` as the return type.
• gotcha Azure Red Hat OpenShift cluster creation typically requires a minimum of 44 vCPUs (or 40 cores in some older documentation) across various VM families. Default Azure subscription quotas are often lower, leading to deployment failures if not increased.
  Fix: Before creating an ARO cluster, request an increase in your Azure subscription's vCPU quota for the relevant VM families (e.g., Standard DSv3/DSv5/Dxv6) through the Azure portal support request.
• gotcha This library (and the broader Azure SDK for Python) requires Python 3.9+. Support for Python 2.7 officially ended on January 01, 2022.
  Fix: Ensure your development and deployment environments are running Python 3.9 or a newer supported version.
• gotcha Certain operation groups (e.g., `OpenShiftClustersOperations`, `OpenShiftVersionsOperations`) should not be instantiated directly. Instead, they are accessed as attributes of the `AzureRedHatOpenShiftClient` instance.
  Fix: Always access operation groups through the client object, e.g., `client.open_shift_clusters.list()` instead of `OpenShiftClustersOperations().list()`.

Install

• pip install azure-mgmt-redhatopenshift azure-identity Install with pip

Imports

• AzureRedHatOpenShiftClient

  from azure.mgmt.redhatopenshift import AzureRedHatOpenShiftClient

• DefaultAzureCredential

  from azure.identity import DefaultAzureCredential

Quickstart

This quickstart demonstrates how to initialize the `AzureRedHatOpenShiftClient` using `DefaultAzureCredential` for authentication. It assumes necessary Azure environment variables (AZURE_CLIENT_ID, AZURE_TENANT_ID, AZURE_CLIENT_SECRET, AZURE_SUBSCRIPTION_ID) are configured. The client can then be used to interact with Azure Red Hat OpenShift resources. An example to list clusters is commented out as it requires specific Azure resources and permissions to run successfully.

import os
from azure.identity import DefaultAzureCredential
from azure.mgmt.redhatopenshift import AzureRedHatOpenShiftClient

# Ensure environment variables are set for authentication:
# AZURE_CLIENT_ID, AZURE_TENANT_ID, AZURE_CLIENT_SECRET
# AZURE_SUBSCRIPTION_ID

# For demonstration, using os.getenv with a placeholder subscription ID.
# In a production environment, ensure these are securely configured.
subscription_id = os.getenv("AZURE_SUBSCRIPTION_ID", "00000000-0000-0000-0000-000000000000")

# Authenticate using DefaultAzureCredential
# This attempts to authenticate via environment variables, managed identity, etc.
credential = DefaultAzureCredential()

# Create a client
client = AzureRedHatOpenShiftClient(credential=credential, subscription_id=subscription_id)

print(f"AzureRedHatOpenShiftClient initialized for subscription: {subscription_id}")

# Example: List OpenShift clusters in the subscription (requires existing clusters and permissions)
# try:
#     for cluster in client.open_shift_clusters.list():
#         print(f"Found cluster: {cluster.name} in resource group: {cluster.resource_group_name}")
# except Exception as e:
#     print(f"Could not list clusters: {e}")

print("To list clusters, ensure you have the necessary permissions and uncomment the example code.")