A Kubernetes Operator for PostgreSQL - Architecture and Design
summary
A Kubernetes operator (pgo) that automates Postgres lifecycle tasks (create, backup, restore, clone, scale, upgrade) using a Go/Cobra CLI and the Kubernetes client API, deployed as a controller that watches third‑party resources → CRDs and materializes services, pods and PVCs. It composes four crunchy container building blocks (crunchy-postgres, crunchy-backup, crunchy-proxy, crunchy-upgrade), requires Kubernetes ≥1.5.3, supports Postgres 9.5/9.6 images, leaves PVCs intact by default, and already performs backup/restore and cloning workflows end‑to‑end via the pgo CLI.
AI-written summary · about
chapters (23)
- 0:00Intro
- 1:15Crunchy Data
- 6:00Operator Basics
- 7:55Why Do I Need This?
- 9:53Building Blocks
- 11:27Installation
- 14:38Installing Kubernetes
- 16:44Create a Database Cluster
- 18:33Showing Cluster Secrets
- 19:23Test a Database
- 19:56Delete a Database Cluster
- 20:45Backup a Database
- 21:55Show Backup PVCS
- 22:44Restore from Backup
- 23:24Show Restored Cluster
- 23:48Scale a Database
- 24:46Perform a Minor Database Upgrade
- 26:14Create a Database Policy
- 27:19Apply a Database Policy
- 28:02Create Policy from GitHub
- 28:41Clone Database
- 29:22Multiple Clusters
- 34:35Strategies (Resources)
related talks
48mPostgreSQL in Kubernetes: Operation - Day 2
50mDemystifying Kubernetes for Postgres DBAs; A Guide to Operators
48mPostgres and Kubernetes: Past, Present and Future
34mRun PostgreSQL:The Kubernetes way
31mLightning Talk: Kubernetes as Your DBA
26mKubernetes from the Database Out
28mLightning Talk: Cloud Native PostgreSQL - Running PostgreSQL on Kubernetes
06mSponsored Keynote: Expanding the Kubernetes Operator Community