0

I have a data science VM(Ubuntu) which is in Corporate Network (No public IP). I'm trying to connect/attach this VM with Azure ML Services(Enterprise).

I get below error Provisioning error : Connection failed: Connection attempt timed out. Verify that server is accessible and SSH service is accepting connections. Resource: '172.16.204.199:8000'. Tried with both ports 8000 and 22 as well.

Client Request ID : 270308d4-1beb-4346-9adc-154307835f1b Service Request ID : |af2659de20174c1d8131d3f05f954c48

It would be great if there's a solution to this. I can confirm that VM is running and I'm able to log into jupyter using the private ip address. Thanks

Sai Krishna Gaddam
  • 1

1 Answers1

0

Unfortunately, you cannot attach a DSVM to an AzureML workspace through a private IP address. You will need a public IP that needs to be reachable from the Azure Machine Learning IP range in the region of the workspace (use an NSG to restrict traffic).

The ranges are published here: https://www.microsoft.com/en-us/download/details.aspx?id=56519

Look for "systemService": "AzureMachineLearning" in the JSON file.

Daniel Schneider
  • 1,797
  • 7
  • 20