Tag: backup

etcd snapshots and Velero on k3s: the native S3 upload flags, the local-path hostPath trap that silently skips your PV data, and a restore drill that works.

pg_dump, pgBackRest, and WAL-G compared for Postgres backups. Point-in-time recovery, S3 setup, retention policy, and the restore drill to run today.

Home Assistant disaster recovery: off-site backups, secrets handling, Z2M coordinator, restore drills that catch missing pieces.

Client-side encryption with rclone crypt keeps your cloud backups opaque to providers. Key management, filename encryption, and restore drills for B2, S3, Drive.

Compare cold storage tiers: Backblaze B2, Storj decentralized cloud, and Scaleway Glacier. Pricing, egress, restore times, and which backup service wins.

Why rsync.net is the no-BS off-site backup target for Restic and Borg: SSH-only ZFS, real Unix support, and what 1.5 cents per GB per month actually costs.

Backup patterns that survive disasters: snapshot, replication, off-site. Retention, safety, and the 3-2-1 model done right.

3-2-1 backups aren't enough anymore. The 3-2-1-1-0 rule adds immutable & offline copies plus verified restores, here's how to implement it.

BorgBackup is great. Borgmatic makes you actually run it: config-driven schedules, hooks, healthchecks, and a backup system that pages you when it breaks.

Your cluster will fail. Velero captures everything, manifests, CRDs, persistent volumes, and lets you replay it on a fresh cluster. Setup, gotchas, real DR plans.

Borg is open-source classic with SSH targets. Duplicacy has lock-free multi-source magic but commercial GUI. Here's which backup tool to pick.

rclone copies and syncs files to cloud storage. Restic does deduplicated encrypted snapshots. Confusing them costs you restores at 2 AM.