Despite best effort and preparation of your precious source code one day you might be facing the situation that there is one or more files in your development workspace that doesn't belong into your git repository.

In general, that's a scenario to use the .gitignore file and exclude such file(s) from being considered or tracked for your repository. This works great but requires that you cover all your tracks regarding potential files, file extensions or folders you don't want to set under version control. Predicting the future is hard.

Plus, changes and audits happen. Meaning, a previously added file should be ignored from now on. Excluding such file in .gitignore won't work as expected out of the box. Such a file needs to be untracked first, and then excluded.

Behold a single one

To untrack a single file that has already been added/initialized to your repository, i.e., stop tracking the file but not delete it from your system use: 

git rm --cached filename

To undo the previous command, add that single file back using git add filename.

Hold my beer!

Occasionally, there is more than a single file to handle. Here's how to untrack every file that is now in your .gitignore and then add all other files back.

First commit any outstanding code changes!

Then, run this command to untrack everything.

git rm -r --cached .

This removes any changed files from the index (staging area).

Afterwards, add everything back into the index, now respecting .gitignore excluding certain files and folders.

git add .

Note: Make sure to commit all your important changes before running git add . Otherwise, you will lose any changes to other files.

Complete the changes by commit.

git commit -m ".gitignore is now working"

Note of caution

Please be careful, when you push this to a repository and pull from somewhere else into a state where those files are still tracked, those files will be deleted.

Image credit: Gemini using prompt create an image showing a junction of railroad tracks