An Azure machine learning service for building and deploying models.
Hello Issac Chan,
Welcome to Microsoft Q&A and Thank you for reaching out.
The issue you encountered is not due to a misconfiguration on your side, and it does not indicate a service outage.
The error:
(AlphanumericVersionNotSupported) Model field AlphanumericVersion is not supported
was caused by a transient issue in the Azure ML Model Registry service, where even valid numeric versions (such as 1 or 0) were being rejected. This behavior was observed more broadly and has since been mitigated.
What happened
The service was incorrectly validating the model version field
Even when using valid numeric values or leaving the version empty, the request failed
This created the appearance of a configuration issue, but it was actually a temporary backend inconsistency
Current status
The issue has now been resolved
Model registration should work as expected moving forward
To avoid similar issues and ensure smooth model registration:
- Use a strictly numeric version (for example:
version: 1) - If you leave the version blank, Azure ML will auto-increment it
- Ensure you are using the latest Azure ML CLI (
az ml) or VS Code extension to pick up recent fixes.
Also please refer this
- Problem with models or model registration: https://docs.microsoft.com/azure/machine-learning/concept-model-management-and-deployment#register-package-and-deploy-models-from-anywhere
- Azure status page: https://azure.microsoft.com/status
I Hope this helps. Do let me know if you have any further queries.
If this answers your query, please do click Accept Answer and Yes for was this answer helpful.
Thank you!