Microsoft.Discovery Supercomputer deployment fails with InvalidResourceType

Question

Title: Microsoft.Discovery Supercomputer deployment fails with InvalidResourceType although provider metadata and ARM validation recognize the resource type

---

Summary

I am attempting to deploy a Microsoft Discovery Supercomputer (Microsoft.Discovery/supercomputers) into subscription 9c69fc18-81c6-4464-8dcc-e0e681724e79.
I have tried to deploy through the Azure portal and through ARM templates and get the same error.

The deployment consistently fails with:

{

"code": "InvalidResourceType",

"message": "The resource type could not be found in namespace 'Microsoft.Discovery' for api version '

Answer 1

Your diagnosis is sound. Provider registration, metadata, and template validation establish that ARM recognizes the namespace and schema; they do not prove the resource provider's runtime route is enabled for a subscription and region. The failed collection GET for every advertised API version rules out a template issue.

There is no customer-side action beyond provider registration that can enable a specific RP route or feature flight. Avoid repeatedly re-registering the provider or changing API versions. Ask the Discovery RP/onboarding team to verify East US availability, subscription feature registration, and the backend route for Microsoft.Discovery/supercomputers, using the tracking ID already captured.

After they confirm the feature is enabled, re-run the same deployment. If it is preview-only or not available in that region, use the region and API version provided by the onboarding team rather than relying on global provider metadata.