Search code examples
debuggingazurefeedbackazure-machine-learning-service

The "perfect error": untraceable, unnamed, from neverland


I have an experiment running without problems when I run single modules as selected parts.

enter image description here

The situation is pretty different when I run the entire experiment. In that case it fails, but I cannot know why.

enter image description here

The experiment returns an error and obviously it doesn't let me deploy the web service, while:

1) I cannot know on which module my error is.

2) I don't have an overall description of the error.

3) It could be related to the error here but I cannot know because I don't have any feedback about that. I know that it could be a bug Azure is trying to solve but this is not reported anywhere.

I really need to know if that's a bug and if I can do something about that.


Solution

  • This issue has been resolved. Please let me know if this happens again