@wikiyu I used to be in the "Look at my insane uptime!" club. I have since learned better.

If any of my servers uptimes are > 31d, I start getting twitchy, because that means monthly patches didn't run like they're supposed to.

@jjbaumgartner Nice point.

About updates I am now on a yolo side of world:

"alias yolo=sudo make all updates on my local pc, on my vps and on my homeserver, dont ask questions. --force && then reboot it"

and beside of that, both - vps and homserver has yolo type update in cron. And then reboot of course

Because - backups are not only to be a backup. I am now really see that I need them in case of issues that may happen at ANY night.

@wikiyu my goal is like a week+ generally - most people just need more ram, think about how much work you can get done in one day, a few minutes of downtime...with all updates i would say rebooting every couple weeks does not have a huge downside, has potential upside #bleeding edge stable

@wikiyu (Sharing a room with a loud IBM Workstation that masqueraded as server was a no go, especially with the girl I used to date, so it had to go to sleep)

@wikiyu Honestly, "server is sleeping" pages deserve to be a standard HTTP status code. 503 Service Unavailable just doesn't capture the vibe.

Also, daily reboots is just aggressive self-healing infrastructure. You're not chaotic, you're ahead of the curve.

@deepjoy

And the most funny part is... that these daily reboots started from weird network errors caused by my previous ISP that was slowing down my internet while connected device has been up for more than 26h.
I was using my server as endpoint for their infrastructure instead of their router (that was rebooting every 12h for some reason).

And that 24h reboot came from that.
Ofcourse i could just restart my network connection. But making reboot was easier. And now its just a routine.

BTW I would say it is should be HTTP 2xx because it is a expected behaviour from my perspective 269 "f off, server went to bed"