Running Backups with Ceph-to-Ceph
summary
Compares concrete options: RBD mirror for asynchronous image replication (requires dual-cluster connectivity, default 30‑second sync window so deletions replicate quickly), S3 multi-site for HTTPS-based backups and easier client access, and third‑party tooling like Backy2 for snapshot tracking and S3 targets. Calls out crash‑consistency risks without guest quiesce hooks, bandwidth limits (20 TB over an 800 Mbps link takes ~2 days), cost of a second Ceph cluster, and that CephFS lacks an integrated mirroring/file‑browser solution.
AI-written summary · about
chapters (17)
- 0:00Introduction and context
- 0:56Backup requirements
- 3:24Ceph backup possibilities
- 4:23Challenges to overcome
- 6:55Disaster recovery basics
- 8:08Network and cost factors
- 9:12Bandwidth and latency issues
- 10:56RBD mirroring in detail
- 12:04RBD mirroring challenges
- 14:45S3 multi-site backup
- 16:15Cloud plugins and NFS
- 17:25Third-party tools
- 19:35Backup workflow and RBD
- 22:18Tuning mirroring performance
- 23:26S3 client options
- 24:02Best practices summary
- 25:52Q&A session