Azure Machine Learning Studio - No Kernel Connected

Question

There is a Kernel Error within Azure Machine Learning Studio.

I have restarted the Compute Instance and used different browsers; the issue does not disappear.

It was working fine until the 31st January 2025. Does anybody know when its going to be fixed?

Answer 1

Hi Gary Cowan !

We have noticed that you rated an answer as not helpful. We appreciate your feedback and are committed to improving your experience with the Q&A.

I'm glad that you were able to resolve your issue and thank you for posting your solution so that others experiencing the same thing can easily reference this! Since the Microsoft Q&A community has a policy that "The question author cannot accept their own answer. They can only accept answers by others ", I'll repost your solution in case you'd like to accept the answer.

Ask: Azure Machine Learning Studio - No Kernel Connected.

Solution: The issue is resolved. that you deleted the compute instance and created a new compute instance in Azure Machine Learning Studio.

If I missed anything please let me know and I'd be happy to add it to this answer, or feel free to comment below with any additional information.

If you have any other questions, please let me know. Thank you again for your time and patience throughout this issue.

 

Please don’t forget to Accept Answer and Yes for "was this answer helpful" wherever the information provided helps you, this can be beneficial to other community members.

Thank You.

Answer 2

Hi Gary Cowan !

Welcome to Azure AI Q and A forum. Thank you for posting your query here.

Kernel error in computes appear when the concerned environment kernel is corrupt or network connection to compute are not happening smoothly (in secured workspace) due to missing/corrupt network configuration.

Here is the troubleshooting procedure for public and private resource.

1. Public workspace. If you have access to computer terminal. 1.1 Verify from Jupyter logs - sudo journalctl -u jupyter for error "jupyter.service: Failed at step CHDIR spawning /anaconda/envs/jupyter_env/bin/jupyter: Permission denied" Please create a new environment kernel and set that as default. Please refer this document.
2. Private workspace. 2.0. Take a network trace (HAR trace) and see for blocked connection which can be whitelisted in your org/azure firewall. (Check 4## and 5## error codes) 2.1 Check DNS resolution of compute for corrupt virtual network links and re-create virtual network links if needed (from a VM in same VNet or terminal) nslookup <ciname>.<region>.instances.azureml.ms -> should resolved to private ip as mentioned in DNS configuration of your private endpoint. 2.2. Check whether you have added "Azuremachinelearing" service tags in outbound rules of NSG on below ports.

     AzureMachineLearning 
   
    443, 8787, 18881 UDP: 5831
   

If the above answer is helpful, please upvote for this answer.

Thank you.

Answer 3

I fixed the issue. I deleted the compute instance and created a new compute instance in Azure Machine Learning Studio