Daniel Barnes

- [x] Amazon Web Services: https://github.com/iterative/minroud-s3 - [ ] Google Cloud - [ ] Microsoft Azure /?

It looks like that app needs an additional scope it might not have? https://docs.github.com/en/rest/actions/workflow-runs#list-workflow-runs-for-a-repository @mikolajpabiszczak to confirm is an issue with app generated token can you try and curl the...

we might need to update our guide for using a github app?  - https://cml.dev/doc/self-hosted-runners#app - https://github.com/settings/apps/new - https://docs.github.com/en/rest/overview/permissions-required-for-github-apps#permission-on-actions - https://docs.github.com/en/rest/actions/workflow-runs#list-workflow-runs-for-a-repository

@mikolajpabiszczak thanks for the report and help, we'll keep this open until we update the docs

Consider using `systemd-run` to start the `run.sh` ? Should scrub vars set in the session with export but maintain system controlled ones like `PATH`

> Consider using `systemd-run` to start the `run.sh` ? Should scrub vars set in the session with export but maintain system controlled ones like `PATH` Ehh don't think this plays...

Confirm that this is really only an issue for cloud-based runners; the fix might be in tpi?

* Related: https://github.com/iterative/terraform-provider-iterative/pull/458

From other experiences with self-hosted runners, this is not really a `cml runner` issue. the only change I would recommend would be to start an idle-timeout check after detecting the...

the crux is that we can't really trust the GH API's job status after something like this.