I ran miniconda, ~/miniconda/bin/conda --help and it took 10 seconds. I hope that’s not normal and am wondering how I might check to see what I’ve done wrong. This is a Linux Mint system with 16 GB memory, a fairly quick ssd, and 6 3.9 GHz GHz processors.
It makes my logins really slow since the bash startup script calls conda twice. I didn’t write it, it was installed when I installed or configured miniconda.
Thanks for any suggestions.
Jim.
Yes, this isn’t normal. Sometimes conda will be slow when run for the first time, but that’s more of a Python problem than a conda problem. When Python programs first run they have to generate a lot of these *.pyc files which takes some time. Subsequent runs are much quicker though.
Is this slow performance consistent for every command invocation and not just the first?
To answer your question, you can add verbose flags to conda (e.g. -v -v -v ), but that won’t show any additional information for conda --help.
The issue is there are a couple of conda commands that are in my .bashrc and are executed every time I start up a terminal session. It takes 15-20 seconds for the terminal to come up. I traced the problem to the conda commands. I don’t know enough about conda to know what they are doing, why they are there and why it’s taking so long. The .bashrc file was modified to include these commands by some part of installing miniconda I think. I didn’t put them there.
It’s normal for Miniconda to edit your bashrc file.
You might want to try setting auto_activate_base to false. Run this command conda config --set auto_activate_base false. This will make it so that conda does not activate upon launching your shell. This is also a debug step to take to make sure it is not coming from conda.
But, just to let you know, this is not normal to have such a slow activation. I wonder what else may be going on on your computer…
Well I don’t know. I’ll try that. Thanks. I don’t use conda on a daily basis so I don’t mind running it manually when the need arises. Python programming is only a small part of what I do.
Thanks,
Jim
1 Like