Endpoint "predict-auto-price" deployment failed

Question

I'm following the "Exercise - Explore regression with Azure Machine Learning designer"

I've redone this exercise twice following all the instructions perfectly!

On the "Deploy a service" section in the exercise, I've tried it over 7 times:

4. In the configuration screen, select Deploy a new real-time endpoint, using the following settings:

Name: predict-auto-price
Description: Auto price regression
Compute type: Azure Container Instance

It always fails with "Endpoint "predict-auto-price" deployment failed"

And nothing comes up in the Deployment Logs.
Please help! I'm out of idea's

""

Answer 1

@Mervyn King Do you see any new jobs added on the jobs tab when your deployment fails? Usually, there is a drop down that loads on the Deployment logs tab to select the Deployment. But, I dont see that on your page. It looks like the page might not have loaded completely. You can also report the issue using the smiley icon on the top right corner with a screen shot of this page.

If an answer is helpful, please click on or upvote which might help other community members reading this thread.

Answer 2

When you are deploying, you must increase CPU reserve capacity to 0.5. I think the default value of 0.1 is not enough for the predictive service deployment.

Answer 3

It seems as if this may either take a longer than expected time to complete or the page will need to be refreshed. I refreshed my page and everything seems to work now with a healthy endpoint state when accessing the Test tab for my endpoint.