We are currently using several containers for our CI/CD pipeline and occasionally running into situation where we have to cancel queued jobs to allow deployment to over them. Adding more containers wouldn’t really help in a long run as we usually have a few jobs sitting in the queue already.
Is there a way to handle this properly? Can we assign a level to each job for prioritization? Or can we expedite deployment job to always go to the front of the queue?
Please feel free to suggest how you are or thinking such circumstances should be handled.