Connection to kernel fails

[raghavchandra123]

Describe the bug
Though working just yesterday, now when trying to run my jupyter notebook, the kernel doesn't connect and the menu only shows default kernel even though I have multiple environments set up and "conda env list" lists all of the environments properly. I have tried restarting everything (jupyter, cpu instance and file) but problem persists

To Reproduce
Steps to reproduce the behavior:
Try to run Jupyter notebook

Expected behavior
Should connect to kernel and run

Screenshots

Desktop (please complete the following information):

• OS: Windows
• Browser Chrome
• Version 108

Additional context
Add any other context about the problem here.

[icoxfog417]

Have you ever connected to the your created conda environment from Jupyter Lab in Studio Lab? When we created conda environment, we should run ipython kernel install to use these as a kernel. It is the default behavior in Jupyter Lab.

[raghavchandra123]

As an update, now I cannot access my files in what seems to be a fatal error and it gives error message: "There was a problem when loading your project. This should be resolved shortly. Please try again later." It would be great if I could somehow download the data or reset the environments if that is what is causing a problem.

[icoxfog417]

As you have already read about the issue, to recreate the account is the solution when we can not launch the project. Unfortunately, the data is cleared when recreating the account.

#138

If you remember what operation caused this error, please let us know to fix the problem that bothering you.

[raghavchandra123]

Yes i have seen this thanks, i am not sure this is a good solution as it results in complete data loss, the operation that caused the error is almost certainly to do with environments and allowing reset to default environment would fix the problem i assume. To me it seems that this is not a well maintained service by amazon, it seems strange that a fatal error reported several times can remain unfixed for so many months with no decent solution or data recovery methods. I will switch over to azure student

[icoxfog417]

Sorry for the inconvenience. If you push the working result to the Git repository, you can clone / pull the contents from it.
We hope you try Studio Lab again after our continuous update.

[raghavchandra123]

Yes but i cant push the result to github without access to the files

[MicheleMonclova]

raghavchandra123, thanks for raising this to our attention. We will see if we can recreate it so it can be fixed.