Azure ML Endpoint Creation Fails: Persistent "SubscriptionNotRegistered" Error on Student Subscription

Question

Hi,

I'm trying to create a Managed Online Endpoint in my Azure Machine Learning workspace (unified-dashboard-ml, region westus2) using an Azure for Students subscription, but it repeatedly fails with the error SubscriptionNotRegistered.

Error Message Snippet:

JSON-
{

"authorization": {

    "action": "Microsoft.MachineLearningServices/workspaces/onlineEndpoints/deployments/getLogs/action",

    "scope": "/subscriptions/REDACTED/resourceGroups/unified-dashboard-rg/providers/Microsoft.MachineLearningServices/workspaces/unified-dashboard-ml/onlineEndpoints/portfolio-prediction-v1/deployments/undefined"

},

"caller": "******@vt.edu",

"channels": "Operation",

"claims": {

    "aud": "https://management.core.windows.net/",

    "iss": "https://sts.windows.net/60956884-10ad-40fa-863d-4f32c1e3a37a/",

    "iat": "1761722103",

    "nbf": "1761722103",

    "exp": "1761726720",

    "http://schemas.microsoft.com/claims/authnclassreference": "1",

    "acrs": "p1",

    "aio": "AQQB+/4aAAAAkH8PIhw5AAgo0e+MaQRkkpdtprrWysd8f9HCiScvbkanYQWDmvNi2D3fKxE7FOpGlgJMSGQDpCkqWVHQNZe6WvvRRrfu23+mSPp7O3PhQp4WrPj3Ru/Lh/z4hMYnaTPEkA1GepwvTe6LWAtoyFWit8lxG9SuTQKUAZfYTGMPPj/I50SmIDpsjsd76b2a71OaquwjtlO+QWHj4i3bCoh/XtMlKHePzy/ySsRyVA1kJQWBu1u83X0bSZpHCrkaHM33SWD6FJ/aR3rgjvEgV0h6blpbpSNFbGRdSjXyIjxxn8QaXvDDYpamBgZC2ghklAPn8AeWITz0PQ8BY7RdNaMD0Q==",

    "http://schemas.microsoft.com/claims/authnmethodsreferences": "fido,mfa",

    "appid": "d7304df8-741f-47d3-9bc2-df0e24e2071f",

    "appidacr": "0",

    "http://schemas.xmlsoap.org/ws/2005/05/identity/claims/surname": "Jain",

    "http://schemas.xmlsoap.org/ws/2005/05/identity/claims/givenname": "Aarav",

    "groups": "0459c90a-da37-4500-a296-1d0b890b6047,45b97516-89f7-4b7d-b6b8-95642e4ca48f,53ee1b5b-7578-45d0-a35c-16a42bc85069,c0f0b48d-7b82-4e9e-afe9-2b8275d6a540,9e7570a6-8b91-4308-a264-b7578ed0a4be,7ea091b7-5f69-433b-a5e7-4ffb4895e25e,59d230e2-aeff-41ee-a56e-2300560d96cf",

    "idtyp": "user",

    "ipaddr": "2607:b400:a00:12:d55f:e763:b888:a453",

    "name": "Jain, Aarav",

    "http://schemas.microsoft.com/identity/claims/objectidentifier": "16c434d3-77bd-4dca-b805-de2bf34eafdd",

    "onprem_sid": "S-1-5-21-1824200278-923733676-1501187911-226349",

    "puid": "1003200485253767",

    "rh": "1.AVsAhGiVYK0Q-kCGPU8yweOjekZIf3kAutdPukPawfj2MBP5AINbAA.",

    "http://schemas.microsoft.com/identity/claims/scope": "user_impersonation",

    "sid": "0089aa49-b031-0f80-ba6a-1af962a96fab",

    "http://schemas.xmlsoap.org/ws/2005/05/identity/claims/nameidentifier": "c2T7EQEBTLHoKF2QAQkOy2ndCp-E6BIR5hvbzpJw9DE",

    "http://schemas.microsoft.com/identity/claims/tenantid": "60956884-10ad-40fa-863d-4f32c1e3a37a",

    "http://schemas.xmlsoap.org/ws/2005/05/identity/claims/name": "******@vt.edu",

    "http://schemas.xmlsoap.org/ws/2005/05/identity/claims/upn": "******@vt.edu",

    "uti": "g4yEEmi_iEKAkGQxHNnfAA",

    "ver": "1.0",

    "wids": "b79fbf4d-3ef9-4689-8143-76b194e85509",

    "xms_act_fct": "3 5",

    "xms_ftd": "1ZPkKFO0o7VeheBPgs_OkjinzAEIrH5r05qm1l0QcbUBdXNlYXN0LWRzbXM",

    "xms_idrel": "1 26",

    "xms_sub_fct": "3 8",

    "xms_tcdt": "1342528483"

},

"correlationId": "56f0a5c6-6dff-4ce9-9a05-59daa724a66f",

"description": "",

"eventDataId": "0f1f4aa3-8fdd-4b4a-84bc-075d4842bf6a",

"eventName": {

    "value": "EndRequest",

    "localizedValue": "End request"

},

"category": {

    "value": "Administrative",

    "localizedValue": "Administrative"

},

"eventTimestamp": "2025-10-29T08:30:48.0722862Z",

"id": "/subscriptions/REDACTED/resourceGroups/unified-dashboard-rg/providers/Microsoft.MachineLearningServices/workspaces/unified-dashboard-ml/onlineEndpoints/portfolio-prediction-v1/deployments/undefined/events/0f1f4aa3-8fdd-4b4a-84bc-075d4842bf6a/ticks/638973234480722862",

"level": "Error",

"operationId": "9e11680f-cceb-4125-993f-6759fe474fd4",

"operationName": {

    "value": "Microsoft.MachineLearningServices/workspaces/onlineEndpoints/deployments/getLogs/action",

    "localizedValue": "Gets deployments Logs in an online inference endpoint in Machine Learning Services Workspace(s)"

},

"resourceGroupName": "unified-dashboard-rg",

"resourceProviderName": {

    "value": "Microsoft.MachineLearningServices",

    "localizedValue": "Microsoft.MachineLearningServices"

},

"resourceType": {

    "value": "Microsoft.MachineLearningServices/workspaces/onlineEndpoints/deployments",

    "localizedValue": "Microsoft.MachineLearningServices/workspaces/onlineEndpoints/deployments"

},

"resourceId": "/subscriptions/REDACTED/resourceGroups/unified-dashboard-rg/providers/Microsoft.MachineLearningServices/workspaces/unified-dashboard-ml/onlineEndpoints/portfolio-prediction-v1/deployments/undefined",

"status": {

    "value": "Failed",

    "localizedValue": "Failed"

},

"subStatus": {

    "value": "NotFound",

    "localizedValue": "Not Found (HTTP Status Code: 404)"

},

"submissionTimestamp": "2025-10-29T08:32:15Z",

"subscriptionId": "b34f90e5-41c0-4670-9132-51d7d309632e",

"tenantId": "60956884-10ad-40fa-863d-4f32c1e3a37a",

"properties": {

    "statusCode": "NotFound",

    "serviceRequestId": null,

    "statusMessage": "{\"error\":{\"code\":\"ResourceNotFound\",\"message\":\"The Resource 'Microsoft.MachineLearningServices/workspaces/unified-dashboard-ml/onlineEndpoints/portfolio-prediction-v1/deployments/undefined' under resource group 'unified-dashboard-rg' was not found. For more details please go to https://aka.ms/ARMResourceNotFoundFix\"}}",

    "eventCategory": "Administrative",

    "entity": "/subscriptions/REDACTED/resourceGroups/unified-dashboard-rg/providers/Microsoft.MachineLearningServices/workspaces/unified-dashboard-ml/onlineEndpoints/portfolio-prediction-v1/deployments/undefined",

    "message": "Microsoft.MachineLearningServices/workspaces/onlineEndpoints/deployments/getLogs/action",

    "hierarchy": "60956884-10ad-40fa-863d-4f32c1e3a37a/b34f90e5-41c0-4670-9132-51d7d309632e"

},

"relatedEvents": []
}

__Troubleshooting Steps Taken:__

1. Verified via the Azure Portal that all required resource providers (`Microsoft.MachineLearningServices`, `Microsoft.Storage`, `Microsoft.Network`, `Microsoft.ContainerRegistry`, `Microsoft.ContainerInstance`, `Microsoft.KeyVault`) show as __'Registered'__ for my subscription (`b34f90e5-41c0-4670-9132-51d7d309632e`).

Used the __"Re-register"__ option for all these providers.

__Waited over 30 minutes__ after re-registering before retrying the endpoint creation.

Tried creating the endpoint via both the Azure ML Studio UI and potentially the CLI

Successfully created a custom Environment using different base images after some trial and error, confirming image pulling works.

The `SubscriptionNotRegistered` error __still occurs__ specifically during the endpoint creation/provisioning phase.

__Latest Failure Details:__

__Correlation ID:__ `[Paste your Correlation ID here]`

__Timestamp (UTC):__ `[Paste your Timestamp here]`

Since I'm on an Azure for Students subscription, I cannot open a formal technical support ticket. Could this be a known issue with student subscriptions, a backend propagation delay, or is there another step I can take?

Thanks for any help!Hi,

I'm trying to create a __Managed Online Endpoint__ in my Azure Machine Learning workspace (`unified-dashboard-ml`, region `westus2`) using an __Azure for Students__ subscription, but it repeatedly fails with the error `SubscriptionNotRegistered`.

__Error Message Snippet:__

JSON


json
{
  

Troubleshooting Steps Taken:

Verified via the Azure Portal that all required resource providers (Microsoft.MachineLearningServices, Microsoft.Storage, Microsoft.Network, Microsoft.ContainerRegistry, Microsoft.ContainerInstance, Microsoft.KeyVault) show as 'Registered' for my subscription (REDACTED).

Used the "Re-register" option for all these providers.

Waited over 30 minutes after re-registering before retrying the endpoint creation.

Tried creating the endpoint via both the Azure ML Studio UI and potentially the CLI (if you did, mention it).

Successfully created a custom Environment using different base images after some trial and error, confirming image pulling works.

The SubscriptionNotRegistered error still occurs specifically during the endpoint creation/provisioning phase.

Latest Failure Details:

• Correlation ID:

    56f0a5c6-6dff-4ce9-9a05-59daa724a66f
  

• Timestamp (UTC):

    2025-10-29T08:30:48.0722862Z
  

Since I'm on an Azure for Students subscription, I cannot open a formal technical support ticket. Could this be a known issue with student subscriptions, a backend propagation delay, or is there another step I can take?

Thanks for any help! Note: Subscription info redacted at support side.

Answer 1

Hi Jain, Aarav

We need these below two service providers to deploy online endpoint.

Microsoft.PolicyInsights and Microsoft.CdnReference - https://learn.microsoft.com/en-us/answers/questions/2262350/cannot-create-managed-online-endpoints-subscriptio

Full list of resource provider permissions.

# Set context
az account set --subscription <subid>

# Core AML prerequisites
az provider register --namespace Microsoft.MachineLearningServices
az provider register --namespace Microsoft.Network
az provider register --namespace Microsoft.ContainerRegistry
az provider register --namespace Microsoft.KeyVault
az provider register --namespace Microsoft.ManagedIdentity
az provider register --namespace Microsoft.OperationalInsights
az provider register --namespace Microsoft.Insights

# Recently seen blockers for managed online endpoints
az provider register --namespace Microsoft.PolicyInsights
az provider register --namespace Microsoft.Cdn

# Optional but commonly needed in end-to-end scenarios
az provider register --namespace Microsoft.Storage
az provider register --namespace Microsoft.Compute
az provider register --namespace Microsoft.Authorization

# Check registration state (Registered expected)
az provider list --query "[?registrationState != 'Registered'].{Namespace:namespace,State:registrationState}" -o table

Please make sure you have owner or contributor or Azure ML data scientist role over AML workspace.

Endpoint and deployment names are explicitly set (undefined as per error log) , Please proofread your yaml files.

Sample endpoint schema.

# endpoint.yml
$schema: https://azuremlschemas.azureedge.net/latest/onlineEndpoint.schema.json
name: portfolio-prediction-v1
auth_mode: key


# deployment.yml
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
name: blue
endpoint_name: portfolio-prediction-v1
model: azureml:<your-registered-model-name>:<version>
environment:
  image: mcr.microsoft.com/azureml/openmpi3.1.2-ubuntu18.04
code_configuration:
  code: ./src
  scoring_script: score.py
instance_type: Standard_F2s_v2
instance_count: 1
``

Please make sure the compute type selected in endpoint schema is available in your subscription.

Please share the error log from your online endpoint as prescribed here if the issue persists.

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

Thank you.