Services: Scheduled service restart

[martastain]

Story

@m-u-r-p-h-y posted:

we should have a possibility to define regular restart of services to keep them alive and prevent any production disturbance.

a typical example is Ftrack leecher stopping processing the events randomly, so having a possibility to restart every day during the night would help to improve stability dramatically...

Additionally, service status only represents the process existence, not that the process is able to process any requests/data

we need some heartbeat check to have more precise status control.

Proposal

Maybe a HEALTHCHECK functionality on the docker level would make sense