Kubernetes from the Database Out
summary
Practical checklist for running Postgres on Kubernetes: demos run on GKE and Picona Everest, networking exposes services as ClusterIP/NodePort/LoadBalancer configured by kube-proxy, and client connections hit a stable service IP. Storage uses PVCs auto-provisioned by CSI and storage classes and must map to independent failure domains. Pods often contain multiple containers (backup sidecar, config sync), here showing four containers and a typical primary plus two replicas layout. Choose between stateful set vs operator approaches; operators automate rolling upgrades and failover while tools like pgBackRest and Patroni determine backup and promotion behavior. Centralize logs off-node for audit.
AI-written summary · about
chapters (7)
- 0:00Introduction and motivation
- 2:31Kubernetes cluster basics
- 7:30Connecting to the database
- 10:42Storage and volumes
- 13:53Pods and containers
- 16:39Operators and custom resources
- 19:54Summary and Q&A
related talks
50mDemystifying Kubernetes for Postgres DBAs; A Guide to Operators
25mPostgres on Kubernetes for the Reluctant DBA
37mData On Kubernetes, Deploying And Running PostgreSQL And Patterns For Databases In a Kubernetes Cluster.
48mPostgreSQL in Kubernetes: Operation - Day 2
31mLightning Talk: Kubernetes as Your DBA
36mKube-native Postgres [I]
48mPostgres and Kubernetes: Past, Present and Future
34mRun PostgreSQL:The Kubernetes way