Azure ML Endpoint Wont deploy on AKS with Azure ML Studio

Question

Got problem while deploying Azure ML real-time Endpoint on AKS with Azure ML Studio. It remains in Transitioning state indefinitely. Looked in AKS event logs and found that error is

Failed to pull image
"viennaglobal.azurecr.io/azureml/azureml_f7815e0137b51ac1464986018a7cb849":
[rpc error: code = Unknown desc = Error response from daemon: Get
https://viennaglobal.azurecr.io/v2/azureml/azureml_f7815e0137b51ac1464986018a7cb849/manifests/latest:
unauthorized: Application not registered with AAD., rpc error: code = Unknown
desc = Error response from daemon: Get
https://viennaglobal.azurecr.io/v2/azureml/azureml_f7815e0137b51ac1464986018a7cb849/manifests/latest:
unauthorized: authentication required, visit https://aka.ms/acr/authorization
for more information.]

It repeat this action with no success. This Inference cluster created with Azure ML Dashboard, and it worked week ago without any unauthorized problems. Tried to create new inference clusters, endpoints but with no success

Answer 1

@Eugene Thanks for the question.This is likely because you don't permission to deploy to AKS. Can you share your role definition?We have forwarded to the product team to check on this.

https://learn.microsoft.com/en-us/azure/machine-learning/how-to-assign-roles#frequently-asked-questions