Environment data
- VS Code version: 1.100.3 (Remote-SSH)
- Jupyter Extension version (available under the Extensions sidebar): 2025.4.1
- Python Extension version (available under the Extensions sidebar): 2025.6.1
- OS (Windows | Mac | Linux distro) and version: macOS (local), Ubuntu (remote)
- Python and/or Anaconda version: Python 3.10.14 (Anaconda)
- Type of virtual environment used (N/A | venv | virtualenv | conda | ...): conda
- Jupyter server running: Remote
Expected behaviour
The embodied_metalearning Jupyter kernel should launch and allow code cells to execute in .ipynb files when connected via VS Code Remote-SSH.
Actual behaviour
The Jupyter kernel fails to start with the error:
SyntaxError: Unexpected end of JSON input.
Only the R kernel is detected, and Python kernels appear missing.
Running Python scripts in the terminal also fails with the same error.
When accessing the same machine via TeamViewer, everything works fine — including running Jupyter Lab manually. And works well when other people ssh to it.
Steps to reproduce:
[NOTE: Self-contained, minimal reproducing code samples are extremely helpful and will expedite addressing your issue]
- SSH into remote machine (
evan@10.81.105.145) via VS Code Remote-SSH
- Activate Conda environment (
conda activate embodied_metalearning)
- Open a Jupyter notebook
- Select kernel
"Python (embodied_metalearning)"
- Attempt to execute a cell
Logs
Output for Jupyter in the Output panel (View→Output, change the drop-down the upper-right of the Output panel to Jupyter)
[warn] Failed to get Python interpreter details from Python Extension API for ~/anaconda3/envs/embodied_metalearning/bin/python SyntaxError: Unexpected end of JSON input
[error] Failed to connect raw kernel session: SyntaxError: Unexpected end of JSON input
[warn] Kernel Error, context = start Error: Failed to start the Kernel 'embodied_metalearning (Python 3.10.14)'
[error] Failed to shutdown kernel... Cannot read properties of undefined (reading 'dispose')
Output for Jupyter in the Output panel (View→Output, change the drop-down the upper-right of the Output panel to Jupyter)
Environment data
Expected behaviour
The
embodied_metalearningJupyter kernel should launch and allow code cells to execute in.ipynbfiles when connected via VS Code Remote-SSH.Actual behaviour
The Jupyter kernel fails to start with the error:
SyntaxError: Unexpected end of JSON input.Only the R kernel is detected, and Python kernels appear missing.
Running Python scripts in the terminal also fails with the same error.
When accessing the same machine via TeamViewer, everything works fine — including running Jupyter Lab manually. And works well when other people ssh to it.
Steps to reproduce:
[NOTE: Self-contained, minimal reproducing code samples are extremely helpful and will expedite addressing your issue]
evan@10.81.105.145) via VS Code Remote-SSHconda activate embodied_metalearning)"Python (embodied_metalearning)"Logs
Output for
Jupyterin theOutputpanel (View→Output, change the drop-down the upper-right of theOutputpanel toJupyter)Jupyterin theOutputpanel (View→Output, change the drop-down the upper-right of theOutputpanel toJupyter)