For someone new to self-hosting, data backup, security, and system stability are the three most critical elements. Generally, it's recommended to automate backups and use RAID or distributed storage to protect against isolated failures. On the security side, certificate management and access logs should be prioritized. For stability, it's worth comparing containerized applications with local hardware and considering a rollback mechanism if necessary. What are your approaches to these?
What are the most reliable approaches when starting with self-hosting?
👁️ 10 views💬 2 replies❤️ 0 likes
2 Replies
Wow bro, I also jumped into the self-hosting adventure recently, and honestly, I really messed up at first when it came to data security. The most critical part I deployed was **immutable backups**—meaning I archive the original data without ever touching it, just keeping it read-only. For example, I take compressed and encrypted backups with `borgbackup` and even send them to AWS S3 while keeping an offline copy. Instead of paying for RAID or distributed storage, I manually sync to 2 different disks (sounds old-school, but I think it’s the simplest and most transparent way).
For security, I set up `certbot` with `nginx` reverse proxy, plus I use fail2ban to protect both ports and SSH from brute force attacks. I send logs to a separate server via `rsyslog` and connect it to Splunk, so now I have a centralized monitoring system independent of any node. For stability, I tried Docker/Kubernetes, but I think it’s a bit overkill—I’m now using manual `systemd` services, which have way less overhead, and rolling back is easy with `systemctl reset-failed`. In short, first priority: don’t lose data, then keep it out of reach, and stability comes last.
What tools do you recommend for automating backups when starting with self-hosting? Is rsync enough, bro, or are you using something else?