Baby Got Back Ups: How to Leverage your CF Blobstore Provider to...
summary
Uses the BBR CLI to back up CF: internal blobstores require local hard links and long drains, external S3 with s3 versioning lets backups be a fast list of keys+version IDs (tiny artifact) but needs lifecycle rules, S3‑compatible without versioning requires a bucket-to-bucket incremental copy that duplicates storage, and GCS currently lacks usable lifecycle behavior so full copies are needed. One customer’s blobstore is dominated by 70–80% droplets, so selective backups (exclude droplets/packages) cut size and downtime but push work to restage apps and lengthen RTO.
AI-written summary · about
chapters (13)
- 0:00Start
- 0:07CLOUD FOUNDRY
- 0:43Introduction
- 1:05Agenda
- 1:39The world today
- 3:22Backup B
- 4:02Stages of BBR backup
- 5:16Desired Goals in Backup Strategies
- 12:54Azure with soft deletes
- 16:07Example Blobstore Composition
- 16:26Selective Backups Motivations
- 17:11Restore Workflow
- 19:01Recap: Blobstores
related talks
30mBurning Down the House: How to Plan for and Deal with Disaster Recovery in CF
18mSecuring S3 Backups Against Ransomware
55mA Solution to the Backup Inconvenience
28mBurning Down The House: How to Plan for and Deal with Disaster Recovery in CF
59mTech deep dive: Backup & on-demand disaster recovery
54mFive hybrid cloud backup and disaster recovery mistakes to avoid
25mData protection design for business continuity
43m50 Shades of Backup