Runners sometimes stuck listening for jobs #102
Description
What happened?
I sometimes notice that runners are stuck listening for jobs.
They must have been registered with Github at some point but when this happens, the runner is no longer listed in the Github UI. I need to manually close the VM and the next VM will start and register correctly.
Does anyone experience something like this?
What are the steps to reproduce?
No idea sadly
What is the expected behavior?
The runner should not get stuck or have some sane timeout behaviour.
Can we implement some form of health check for runners that have not received a job for a while?