> How many 9s are you getting in your closet at home? Excluding power outages that are out of my control, during which I couldn't do any work anyway, I've had

> How many 9s are you getting in your closet at home? Excluding power outages that are out of my control, during which I couldn't do any work anyway, I've had practically zero downtime of my homelab services in the many years I've been running them. Even the dozen hard drives in my NAS with several years of power on time have given me zero issues. I know that I've been lucky, and according to SnapRAID the probability of one failing in the next year is 82%, but so far I haven't had a failure yet. Even when it does happen, I'm fairly confident that the interruption will be minimal and my data will be safe. All this is to say that running and managing services yourself doesn't require much effort at all, assuming that you're technically inclined and enjoy tinkering. The idea that cloud services are inherently more robust is a myth. Most service interruptions happen because of two reasons: large scale distributed systems are very difficult to run and maintain, and the constant churn of large engineering teams introduces many operational risks. Essentially, it's all due to complexities of scale. These are not problems that a machine in a closet serving a few users will ever have, especially if you're smart about choosing simple and robust technology.