Add correct commands for steps on Creating a Docker container action page

[avinal]

Code of Conduct

• I have read and agree to the GitHub Docs project's Code of Conduct

What article on docs.github.com is affected?

https://docs.github.com/en/actions/creating-actions/creating-a-docker-container-action

What part(s) of the article would you like to see updated?

The Parts that need modification

• https://docs.github.com/en/actions/creating-actions/creating-a-docker-container-action#writing-the-action-code
• https://docs.github.com/en/actions/creating-actions/creating-a-docker-container-action#commit-tag-and-push-your-action-to-github

Context

Seven months ago I worked on an issue #974 and submitted PR #1256 and #2988. There was performance concern over making shell files executable explicitly inside the Docker. This
#1256 (comment) has all the discussion.

Ultimately it was decided to suggest the users to run chmod +x entrypoint.sh on their machine before pushing it to the remote. I was not convinced that this will work and there won't be any permission issue while running the script inside GitHub Actions and I am still not convinced.

Root of the problem.

For the most part, the above method works fine. The problem occurs only when someone clones the repo and then make changes. The permission is reset and it no longer can be run inside GitHub Actions. And that is why you will see a lot of developers/organizations have explicitly made the shell executable inside GitHub Actions before running it.

What do I suggest

Git provides a way to explicitly change the permission mode of a file so that it doesn't get reset every time there is a clone/fork. The command works as below.

 git update-index --chmod=+x entrypoint.sh

This will change the permission mode of the file in the git index and it can be checked by running following command.

git ls-files --stage entrypoint.sh
# output
100755 e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 0       entrypoint.sh

Here 755 denotes the executable permission.

Additional information

These lines here https://github.com/github/docs/blame/main/content/actions/creating-actions/creating-a-docker-container-action.md#L124-L128 have no use and can be removed if we opt for the above solution.

[welcome]

Thanks for opening this issue. A GitHub docs team member should be by to give feedback soon. In the meantime, please check out the contributing guidelines.

[ramyaparimi]

@avinal
Thanks so much for opening an issue! I'll triage this for the team to take a look 👀

[Kellyjo152]

Code of Conduct

• I have read and agree to the GitHub Docs project's Code of Conduct

What article on docs.github.com is affected?

https://docs.github.com/en/actions/creating-actions/creating-a-docker-container-action

What part(s) of the article would you like to see updated?

The Parts that need modification

• https://docs.github.com/en/actions/creating-actions/creating-a-docker-container-action#writing-the-action-code
• https://docs.github.com/en/actions/creating-actions/creating-a-docker-container-action#commit-tag-and-push-your-action-to-github

Context

Seven months ago I worked on an issue #974 and submitted PR #1256 and #2988. There was performance concern over making shell files executable explicitly inside the Docker. This
#1256 (comment) has all the discussion.

Ultimately it was decided to suggest the users to run chmod +x entrypoint.sh on their machine before pushing it to the remote. I was not convinced that this will work and there won't be any permission issue while running the script inside GitHub Actions and I am still not convinced.

Root of the problem.

For the most part, the above method works fine. The problem occurs only when someone clones the repo and then make changes. The permission is reset and it no longer can be run inside GitHub Actions. And that is why you will see a lot of developers/organizations have explicitly made the shell executable inside GitHub Actions before running it.

What do I suggest

Git provides a way to explicitly change the permission mode of a file so that it doesn't get reset every time there is a clone/fork. The command works as below.

 git update-index --chmod=+x entrypoint.sh

This will change the permission mode of the file in the git index and it can be checked by running following command.

git ls-files --stage entrypoint.sh
# output
100755 e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 0       entrypoint.sh

Here 755 denotes the executable permission.

Additional information

These lines here https://github.com/github/docs/blame/main/content/actions/creating-actions/creating-a-docker-container-action.md#L124-L128 have no use and can be removed if we opt for the above solution.

[skedwards88]

Thank you for this detailed issue! You or anyone else is welcome to open a PR to address this.

[avinal]

Thanks @skedwards88, I would like to open a PR.