Besides “don’t use Cloudflare/AWS/etc”, how can we make our selfhosted setups resilient to outages like the ones we’ve seen recently?
You must log in or # to comment.
Consider if you need three/five nines uptime. I know I don’t so I don’t worry if jellyfin or tickdone are down for a few hours.
I think it’s more just in general and not specifically because they need it.
How much money are you willing to spend? Resiliency is expensive.
Implement fall backs. If your selfhosted services are public and mission critical, you should have something in your trick bag to fall back to.
Either figure out a way to multi-cloud host, or just don’t use them, because they’re a single point of failure.
Reduce the number of single failure points. How you choose to do that is up to you and what you can afford.
