Deploying my model to ACI takes forever and fails without any error message. In the ML workspace, the status of the deployed endpoint is unhealthy. I checked common errors while deployment but could not solve the problem. Pleas help. The deployment is never successful and it keeps running.
@lakpatamang-3662 Thanks for the question. Could you clarify the architecture of your model deployment? In particular, are you using a custom docker container for it? Also, usually ACI would be used for testing, but I'd recommend investigating AKS for production model deployment.
I would deploy the container into a local machine/VM with Docker to see the exact detail error message which you don't see via ACI deployment.
10 people are following this question.
