VSCode changed "conda activate [environment]" to "source activate [environment]" in remote ssh

[ramajoballester]

Type: Bug

Behaviour

Expected vs. Actual

The problem I'm facing is that vscode executes "source activate [environment]" instead of "conda activate [environment]" only when I connect to my server via ssh. My server runs ubuntu 20.04 and same thing happens when connecting from windows or linux. It always worked well until last afternoon. I tried updating vscode, but it didn't solve it.

Steps to reproduce:

As I've always done, selecting my python interpreter from my conda env in local works well, but it does not when I do it in remote.

Diagnostic data

• Python version (& distribution if applicable, e.g. Anaconda): 3.8.16
• Type of virtual environment used (e.g. conda, venv, virtualenv, etc.): Conda
• Value of the python.languageServer setting: Default

Output for Python in the Output panel (View→Output, change the drop-down the upper-right of the Output panel to Python)

XXX

User Settings

languageServer: "Pylance"

Extension version: 2022.20.2
VS Code version: Code 1.74.3 (97dec172d3256f8ca4bfb2143f3f76b503ca0534, 2023-01-09T16:57:40.428Z)
OS version: Linux x64 5.15.0-58-generic
Modes:
Sandboxed: No
Remote OS version: Linux x64 5.15.0-58-generic

System Info

Item	Value
CPUs	Intel(R) Core(TM) i7-6500U CPU @ 2.50GHz (4 x 2600)
GPU Status	2d_canvas: enabled canvas_oop_rasterization: disabled_off direct_rendering_display_compositor: disabled_off_ok gpu_compositing: enabled multiple_raster_threads: enabled_on opengl: enabled_on rasterization: enabled raw_draw: disabled_off_ok skia_renderer: enabled_on video_decode: disabled_software video_encode: disabled_software vulkan: disabled_off webgl: enabled webgl2: enabled webgpu: disabled_off
Load (avg)	1, 1, 1
Memory (System)	7.64GB (4.31GB free)
Process Argv	--unity-launch --crash-reporter-id bd1ec8d7-9008-4608-8289-77381a821054
Screen Reader	no
VM	0%
DESKTOP_SESSION	ubuntu
XDG_CURRENT_DESKTOP	Unity
XDG_SESSION_DESKTOP	ubuntu
XDG_SESSION_TYPE	x11

Item	Value
Remote	SSH: uc3m_server
OS	Linux x64 5.15.0-58-generic
CPUs	AMD Ryzen 7 3700X 8-Core Processor (16 x 2200)
Memory (System)	62.75GB (59.62GB free)
VM	0%

A/B Experiments

vsliv368:30146709
vsreu685:30147344
python383cf:30185419
vspor879:30202332
vspor708:30202333
vspor363:30204092
vslsvsres303:30308271
pythonvspyl392:30443607
vserr242:30382549
pythontb:30283811
vsjup518:30340749
pythonptprofiler:30281270
vsdfh931cf:30280410
vshan820:30294714
vstes263:30335439
vscorecescf:30445987
pythondataviewer:30285071
vscod805:30301674
binariesv615:30325510
bridge0708:30335490
bridge0723:30353136
cmake_vspar411:30581797
vsaa593cf:30376535
pythonvs932:30410667
cppdebug:30492333
vsclangdc:30486549
c4g48928:30535728
dsvsc012cf:30540253
azure-dev_surveyone:30548225
vsccc:30610678
pyindex848cf:30577861
nodejswelcome1:30587005
3biah626:30602489
f6dab269:30613381

[karrtikr]

Should be fixed with #11039 once we start using conda run for activation instead. Meanwhile, you can make the source activate <environment> command work by putting activate in PATH.

Make sure $HOME/<conda>/bin or $HOME/<conda>/condabin is in PATH.

In case that doesn't fix the issue, can you provide a screenshot of the error you're getting when source activate <env> is run? Please also provide the location of conda installation.

[ramajoballester]

$HOME/anaconda3/condabin was already in PATH. source activate <env> throws error bash: activate: no such file or directory. Nonetheless, if I run conda activate or conda activate <any-env> I can then run source activate <env>.

However, the question I would like to answer is: is there a way to revert the initial command that the terminal executes when I launch it from source activate <env> to conda activate <env> and why it changed recently? It has always been with conda instead of source and I would like to keep it that way. In addition, this is the way it works on my local pc (with conda activate <env>) but it differs when I connect to my server via ssh, which, apparently, has the same ubuntu, conda and configuration as my local vscode.

[karrtikr]

is there a way to revert the initial command that the terminal executes

Unfortunately no.

I'm not sure why it changed. All this activation code will go away with #11039, so instead of fixing the issue I was hoping we would make the source activate <env> command work for you instead.

$HOME/anaconda3/condabin was already in PATH. source activate throws error bash: activate: no such file or directory

What about $HOME/<conda>/bin? It should probably contain the activate script.

[karrtikr]

Also, what is the version of conda you're using?

[ramajoballester]

Okay, I can confirm that adding $HOME/<conda>/bin to PATH allows me to run source activate <env>.

I just found out the root of the problem. With conda==23.1.0 VSCode runs source activate <env> by default. After downgrading it to conda==22.11.1 it defaults to conda activate <env>.

Thank you so much and let me know if you want any other info related to this issue before closing it.

[karrtikr]

Interesting, well glad you have a workaround for now. I'm closing this issue as we'll be rewriting activation with #11039.

[gjoseph92]

After upgrading conda, I've been having the same issue reported here. I'm using local environments on macOS.

After downgrading it to conda==22.11.1 it defaults to conda activate <env>.

What was the reasoning for this? I'm not aware of source activate <env> ever being the proper way to activate a conda environment, regardless of version, so why is this behavior dependent on the conda version?

I can confirm that downgrading conda to 22.11.1 also causes VSCode to use the correct command, and resolves the issue for me.