Lol Funny you mention that, I used to get a little tingle when I'd see how long a something had been up. I'm happy about the phones, but it is sad on the server side. Especially when you consider all the bullshit that goes into that stuff at scale. The people running CentOS got rugged on top of that.

I think that's changed a lot. I had an ubuntu 14 (iirc) That had over 1000 days of uptime at one point. That wasn't smart, it was because I was lazy and it was a backup machine sitting in a remote location. High uptime is now pretty correlated with low security. I think security-first has shifted heavily since RedHat started to run with it. That means rapid kernel patches, which means package rebuilds and rapid merging. I think leading to this more 3-6 month window like @Matt 🛸 suggested as well of stability. You just have to keep jumping if you want to stay on top of it imo. Consumer cares more about stability, commercial cares more about 99% forever. If 99.9999 is a goal, but one bug in a package takes you down, it wasn't worth it.