Dependency removed: Any single copy of your data or single instance of a service - no individual failure causes permanent loss.
Redundancy is the practice of maintaining multiple independent copies, paths, or instances of something critical so that no single failure causes unacceptable loss. The 3-2-1 rule is the foundation: 3 copies, 2 different media types, 1 off-site.
Redundancy applies to data, services, hardware, and communications. A resilient stack fails gracefully rather than catastrophically.
- 3-2-1 rule - 3 copies of data; 2 different media (e.g., NAS + external drive); 1 off-site (encrypted cloud or physical remote)
- 3-2-1-1-0 rule - adds 1 immutable (air-gapped or WORM) copy and 0 errors on verified restores
- Automated backups - Restic, BorgBackup, or Duplicati; automated schedules; encrypted at rest and in transit
- Versioned backups - retaining multiple versions to recover from ransomware or accidental deletion
- Geographic redundancy - keeping a copy in a physically different location; family member's house; safety deposit box
- Offline backups - air-gapped drives that cannot be ransomwared; disconnected after each backup job
- Service failover - automatic or manual switch to a backup instance when the primary fails; tested before needed
- High availability - clustering and replication for services where brief downtime is unacceptable
- Replication - Nextcloud to a second server; PostgreSQL streaming replication; ZFS send/receive
- Recovery testing - restores must be tested regularly; an untested backup is not a backup
- Identity documents (passports, licenses, certificates) - encrypted, at least 3 locations
- Financial records and credentials - encrypted; paper backup of master password
- Irreplaceable personal data (photos, personal writing)
- System configurations and infrastructure-as-code
- Application data (Nextcloud, Gitea, Wiki)
- Media and reference libraries (lower priority; replaceable with effort)
- Three-Site Backup Architecture - local NAS + off-site encrypted drive + encrypted cloud; automated with Restic or Borg
- Household Disaster Recovery Plan - documented recovery procedures for every critical service; tested quarterly