Personally? Months. Regularly weeks. About the same as my servers. Uptime on a single machine isn’t a metric of anything meaningful.

That said, GitHub ain’t a single machine and the reliability issues are definitely not a good look.

Wait, that’s like C with extra fewer steps

ADHD, eh?

Also, when you start hosting more than 2-3 services, keeping them up-to-date might become a hassle which you can forget. For docker-based hosting I’d recommend you setup watchtower, which can keep your servives up-to-date for you.

I would caution against automatic updates! Notifications, yes absolutely. But automatically updating things is a great way to have things break suddenly when you’re not in a good place to troubleshoot.

This might be a little more than you’re asking for, but I would recommend something like Node-RED with a feedparser. Reason being: it opens up a lot of options and could be a useful tool for things other than this one use case.