Docker is not made that way so people can get out of fixing shit that breaks constantly. The point is more to get rid of once-in-a-lifetime errors for which a reinstall or restart would be faster for anyway, but the main reason is that you get to sleep because if something fucky happens at 3AM, it’s more likely the system will just right itself.
I certainly wouldn’t call debugging a blinking container easier than just fixing shit on the server. In a container though, I can be sure the issue is fixed and it’s not just another bandaid on a growing pile that is going to explode as soon as the guy building the pile retires.
Docker is not made that way so people can get out of fixing shit that breaks constantly. The point is more to get rid of once-in-a-lifetime errors for which a reinstall or restart would be faster for anyway, but the main reason is that you get to sleep because if something fucky happens at 3AM, it’s more likely the system will just right itself.
I certainly wouldn’t call debugging a blinking container easier than just fixing shit on the server. In a container though, I can be sure the issue is fixed and it’s not just another bandaid on a growing pile that is going to explode as soon as the guy building the pile retires.