Unable to Create Online Endpoint in Azure ML — Resource Provider Error

Question

Unable to Create Online Endpoint in Azure ML — Resource Provider Error

Manuel Gaunaurd 20

Hi all,

I’m trying to deploy an Azure Machine Learning online endpoint, and I’m running into this error both in the Azure CLI and directly in the Azure Portal:

(SubscriptionNotRegistered) Resource provider [N/A] isn't registered with Subscription [N/A].

Please see troubleshooting guide, available here: https://aka.ms/register-resource-provider

Code: SubscriptionNotRegistered

What I’ve already done:

Verified that Microsoft.MachineLearningServices is registered.

Noticed Microsoft.Network wasn’t registered initially — I’ve registered it now.

Checked that my subscription is active, enabled, and correctly set as default (az account show confirms this).

Tried deleting and recreating the endpoint with a new name — same issue.

Error message provides no detail on which specific provider is unregistered — shows [N/A] for both subscription and provider.

This issue happens even in the portal UI, so it doesn’t appear to be a CLI-specific configuration problem.

Has anyone experienced this? What other resource providers need to be registered for ML online endpoints to work? Or is this an issue on Azure’s backend?

Thanks in advance — appreciate any help!

Manas Mohanty 5,700 Reputation points Microsoft External Staff Moderator

2025-06-20T17:23:47.1966667+00:00

Hi Manuel Gaunaurd

Please let us know if you were able to deploy the endpoints after registering above providers.

Thank you.
Manuel Gaunaurd 20 Reputation points

2025-06-20T21:12:55.55+00:00

From the first three resource providers, Microsoft.CDN was the only one not registered.

Now when I create an endpoint, it says succeeded in the provisioning state under the endpoint attributes section. However, in the deployment section it says provisioning state failed. It gives the following error:

 ResourceOperationFailure: User container has crashed or terminated. Please see troubleshooting guide, available here: https://aka.ms/oe-tsg#error-resourcenotready 
Manas Mohanty 5,700 Reputation points Microsoft External Staff Moderator

2025-06-20T21:44:32.94+00:00
Hi Manuel Gaunaurd

Thank you for acknowleding my inputs.

ResourceOperationFailure: User container has crashed or terminated. Please see troubleshooting guide, available here:

Above error suggests environment might have crashed. You can check into environment logs to find which was one exact culprit among below possible situation

Scoring script - not able to load model, import package etc - You need to use try-catch block to find the responsible failing code block

Deployment configuration - timeout, liveness, retry etc - can be catched from environment logs

Missing dependency etc. (you need to unpin few versions of packages)

Increase size of VM and

reduce overall package size in requirement.txt or conda.yaml to avoid bulky environment creates

Helpful resource on debugging scoring scripts - https://learn.microsoft.com/en-us/azure/machine-learning/how-to-inference-server-http?view=azureml-api-2#typeerror-during-inference-server-startup

Reference used - https://learn.microsoft.com/en-us/azure/machine-learning/how-to-troubleshoot-online-endpoints?view=azureml-api-2&tabs=cli#error-resourcenotready

Please take a moment to accept the answer on reported scope of the case and create a separate thread to investigate container crashed issue.

Thank you

Accepted answer

0 additional answers

Your answer

Manas Mohanty 5,700 Reputation points Microsoft External Staff Moderator

2025-06-20T17:23:47.1966667+00:00

Hi Manuel Gaunaurd

Please let us know if you were able to deploy the endpoints after registering above providers.

Thank you.
Manuel Gaunaurd 20 Reputation points

2025-06-20T21:12:55.55+00:00

From the first three resource providers, Microsoft.CDN was the only one not registered.

Now when I create an endpoint, it says succeeded in the provisioning state under the endpoint attributes section. However, in the deployment section it says provisioning state failed. It gives the following error:

 ResourceOperationFailure: User container has crashed or terminated. Please see troubleshooting guide, available here: https://aka.ms/oe-tsg#error-resourcenotready 
Manas Mohanty 5,700 Reputation points Microsoft External Staff Moderator

2025-06-20T21:44:32.94+00:00

Hi Manuel Gaunaurd

Thank you for acknowleding my inputs.

ResourceOperationFailure: User container has crashed or terminated. Please see troubleshooting guide, available here:

Above error suggests environment might have crashed. You can check into environment logs to find which was one exact culprit among below possible situation

Scoring script - not able to load model, import package etc - You need to use try-catch block to find the responsible failing code block

Deployment configuration - timeout, liveness, retry etc - can be catched from environment logs

Missing dependency etc. (you need to unpin few versions of packages)

Increase size of VM and

reduce overall package size in requirement.txt or conda.yaml to avoid bulky environment creates

Helpful resource on debugging scoring scripts - https://learn.microsoft.com/en-us/azure/machine-learning/how-to-inference-server-http?view=azureml-api-2#typeerror-during-inference-server-startup

Reference used - https://learn.microsoft.com/en-us/azure/machine-learning/how-to-troubleshoot-online-endpoints?view=azureml-api-2&tabs=cli#error-resourcenotready

Please take a moment to accept the answer on reported scope of the case and create a separate thread to investigate container crashed issue.

Thank you

Answer 1

Hi Manuel Gaunaurd

Please force register all necessary resource providers in your subscription

Microsoft.MachineLearningServices: Required for creating the Azure Machine Learning workspace.
Microsoft.CDN
Microsoft.PolicyInsights

(please start with above 3 provider and add below ones gradually)
Microsoft.Storage: Used as the default storage for the workspace.
Microsoft.ContainerRegistry: Used to build Docker images.
Microsoft.KeyVault: For storing secrets.
Microsoft.Notebooks: For integrated notebooks on Azure Machine Learning compute instance.
Microsoft.Network (If virtual network)
Microsoft.ContainerService: If deploying trained models to Azure Kubernetes Services.

Reference -

https://learn.microsoft.com/en-us/answers/questions/2265489/how-to-fix-resourceoperationfailure-error-on-azure

https://learn.microsoft.com/en-us/answers/questions/2263599/i-try-to-deploy-ml-end-point-and-i-keep-getting-an

Thank you.

Share via

Unable to Create Online Endpoint in Azure ML — Resource Provider Error

0 additional answers

Your answer