Open TheFrenchGhosty opened 3 years ago
Great idea. Seems like a simple patch to a common and known problem.
Wouldn't it be easier and/or better to diagnose what's provoking these crashes, and try to fix it?
@SamantazFox It would be better, yes. However this issue has been a thing for years and it was never fixed.
Supposedly it's related to lsquic however everytime we tried to update it, it caused more issues... so yeah.
Just a suggestion.Can we use something like docker's healthcheck to accomplish this target: like
invidious:
healthcheck:
test: "kill 1"
start_period: 1h ## New in compose version 3.4
interval: 1h
It's an hacky suggestion though. I would prefer to introduce a solution directly implemented into the docker container.
This sounds simple. Why is it still not implemented?
Is your enhancement request related to a problem? Please describe.
Invidious crashes for no reason.
Describe the solution you'd like
A cronjob that restart it, build into the container, to ease deployment, and allows most "amateur" admins to host instances that don't break.
Describe alternatives you've considered
External restart: but it's not ideal.
Additional context