microsoft / vscode-tools-for-ai

Azure Machine Learning for Visual Studio Code, previously called Visual Studio Code Tools for AI, is an extension to easily build, train, and deploy machine learning models to the cloud or the edge with Azure Machine Learning service.
Other
324 stars 94 forks source link

authentication error when connecting to AML CI with VS code desktop & web #2350

Closed lybrian closed 1 month ago

lybrian commented 5 months ago

Does this occur consistently%3F Repro steps:

1. 2.

Error message: Failed to authenticate. Error type: AuthError

Version: 0.48.0 Browser Name: Microsoft Edge Browser Version: 123.0.0.0 OS: Windows 10 64-bit Product: Visual Studio Code Product Version: 1.88.1 Language: en

Call Stack ``` kt.getSession webOpener.js:2:706485 https:/webOpener.js:2:785853 Ar.measure webOpener.js:2:792927 Tr.getSession webOpener.js:2:785811 Tr.create webOpener.js:2:785519 $n.resolve webOpener.js:2:742087 https:/webOpener.js:2:772433 Ar.measure webOpener.js:2:792927 Lr.doRoute webOpener.js:2:772390 Module.Ir webOpener.js:2:793530 ```
sevillal commented 5 months ago

@lybrian thanks for filing this. This is a known issue that we are actively working with VS Code to get fixed, I will post next steps later this week. For reference this is the VS Code upstream issue

sevillal commented 3 months ago

Hi @lybrian, VS Code has implemented a solution to this issue, as of now we are testing it, could you please try it and report if your issue is resolved:

  1. Go to Azure ML Studio and open the Compute tab
  2. Right click on the VS Code (Web) link of your compute instance and click Copy link
  3. Open a new browser tab
  4. Paste the link and add &useMsal=true at the end of the URL
  5. Hit enter
  6. Enter your credentials if needed and report if you are able to connect or if you see a different error
sevillal commented 1 month ago

Hi @lybrian , we just rolled out a fix for this issue, could you please retry? Please reopen this issue if you face any problem.