While there's an existing retry mechanism for the `docker pull` command
[^1], it's missing for `docker login`.
Similar to the `docker pull` scenario, the container registry could
potentially be briefly unavailable or inaccessible, leading to failed
`docker login` attempt and subsequent workflow run failures.
Since it's container based workflow, there is not way to retry on
customer side. The runner should retry itself.
It also aligns with community feedback [^2].
[^1]: 8e0cd36cd8/src/Runner.Worker/ContainerOperationProvider.cs (L201)
[^2]: https://github.com/orgs/community/discussions/73069
Co-authored-by: Thomas Boop <52323235+thboop@users.noreply.github.com>
GitHub Actions Runner
The runner is the application that runs a job from a GitHub Actions workflow. It is used by GitHub Actions in the hosted virtual environments, or you can self-host the runner in your own environment.
Get Started
For more information about installing and using self-hosted runners, see Adding self-hosted runners and Using self-hosted runners in a workflow
Runner releases:
Contribute
We accept contributions in the form of issues and pull requests. The runner typically requires changes across the entire system and we aim for issues in the runner to be entirely self contained and fixable here. Therefore, we will primarily handle bug issues opened in this repo and we kindly request you to create all feature and enhancement requests on the GitHub Feedback page. Read more about our guidelines here before contributing.



