This could be caused by OOM or an application panic. For the latter, usually checking the previous container's logs may
help you quickly find the panic causing the problem:
kc logs --previous -n prod <podname>
To check why the restarts are happening in the first place:
kc describe pod -n prod <podname>
Then look for: Containers -> Last State -> Reason