On a remote server, kernel picker doesn't show virtual environments

Issue Type: Bug

I use the Remote - SSH and Python extensions to run Python code on a remote server. I created a virtual environment test_venvon the remote server I connect to, which contains pandas and jupyter. I’m able to select this environment in VS Code:

I can run a simple Python script from VS Code using this interpreter:

import pandas as pd
filename = 'test.csv'
df = pd.read_csv(filename)
print(df)

This runs fine. However, if I try to run a Jupyter notebook which does the same things:

{
 "metadata": {
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": 3
  },
  "orig_nbformat": 2
 },
 "nbformat": 4,
 "nbformat_minor": 2,
 "cells": [
  {
   "source": [
    "import pandas as pd"
   ],
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'test.csv'\n",
    "df = pd.read_csv(filename)\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ]
}

VS Code complains that ipykernel is not installed, and it doesn’t run the notebook:

This means that VS Code uses the selected environment when running a Python script, but not when running a notebook, as also confirmed by the fact that it’s trying to use Python 2.7 (the virtual environment has Python 3.8) . This is a major bug: basically you cannot run Jupyter notebooks with VS Code on remote servers.

Extension version: 2021.5.842923320 VS Code version: Code 1.56.2 (054a9295330880ed74ceaedda236253b4f39a335, 2021-05-12T17:44:30.902Z) OS version: Darwin x64 19.6.0 Remote OS version: Linux x64 5.4.0-52-generic

System Info

Item	Value
CPUs	Intel® Core™ i7-7820HQ CPU @ 2.90GHz (8 x 2900)
GPU Status	2d_canvas: enabled gpu_compositing: enabled metal: disabled_off multiple_raster_threads: enabled_on oop_rasterization: enabled opengl: enabled_on rasterization: enabled skia_renderer: disabled_off_ok video_decode: enabled webgl: enabled webgl2: enabled
Load (avg)	2, 2, 2
Memory (System)	16.00GB (0.87GB free)
Process Argv	–crash-reporter-id df4f7917-f728-4290-9500-a3632318c3e7
Screen Reader	no
VM	0%

Item	Value
Remote	SSH: xxxx
OS	Linux x64 5.4.0-52-generic
CPUs	Intel® Xeon® CPU E5-2698 v4 @ 2.20GHz (80 x 2696)
Memory (System)	503.78GB (343.49GB free)
VM	0%

[DavidKutu]

I have another couple of questions/suggestions @AndreaPi.

• Could you please update the Jupyter release to the latest version, and check if the issue still happens?
• The python file and the notebook you’re running, are they on the same folder?

[AndreaPi]

Installing the new Jupyter release (which just recently came out) solved everything! Thank you @DavidKutu . Closing the issue now