Rook Deep Dive
summary
Rook claims to orchestrate distributed storage (Ceph, Minio) inside Kubernetes using CustomResourceDefinitions and an operator kit that uses informers and clientsets; example shows a Minio operator built as an Ubuntu xenial container that pulls a Minio binary and the Rook binary. A user-grade cluster runs on an 8-node cluster (i7, 32 GB RAM, 2×4 TB disks) delivering ~46–57 TB usable, hosting Postgres (Stolon), GitLab, Nextcloud, Plex, Elasticsearch, with Prometheus/Grafana monitoring and scrubbing alerts; performance is reported comparable to bare metal and Rook itself stays out of the data path.
AI-written summary · about
chapters (16)
- 0:00Intro
- 0:21What is Rook
- 2:02Rook Challenges
- 3:01Solutions
- 4:39Operators
- 5:11Rook Framework
- 5:40Implementing an Operator
- 13:04Helpful Tips
- 15:11Specifications
- 16:33Disk Player
- 19:09Applications
- 23:18Scaling
- 26:30Mining
- 27:12Elastic Search
- 29:49Summary
- 31:14Questions
