<!--IssueSummary start--> <details> <summary> Everyone can contribute. [Help move this issue forward](https://handbook.gitlab.com/handbook/marketing/developer-relations/contributor-success/community-contributors-workflows/#contributor-links) while earning points, leveling up and collecting rewards. </summary> - [Close this issue](https://contributors.gitlab.com/manage-issue?action=close&projectId=278964&issueIid=15953) </details> <!--IssueSummary end--> ### Description I know it's possible to replicate the docker based CI steps locally, but if that involves a lot of steps or many variables (sometimes secret) it can be a bit complex. I've seen other systems allow you a shell into a build. The use-case being mainly for debugging failed builds. ### Proposal Next to a failed build, we could have a "retry with shell access". Once a digitalocean instance is provisioned for this (I'm assuming gitlab.com here) it could provide SSH access and keep the instance alive for a maximum amount of time to debug the problem. This could increase the digital ocean resources but it could also stop repeated commits and builds with minor changes, each time meaning a new DO instance is created per commit. Ideally the environment variables would also be set in this shell. ### Links / references - https://scrutinizer-ci.com/docs/troubleshooting/ssh_access_to_build - https://blog.codeship.com/codeship-ssh-debug/