KubeArchive: Don’t use Kubernetes as a Database
summary
KubeArchive streams Kubernetes objects to a real database (demo uses Postgres) in real time via KNative Eventing API ServerSource, evaluates user-provided CEL expressions to decide when to write or when to delete, and issues Kubernetes API deletes after successful archive. A REST API exposes archived objects using the same endpoints as the Kubernetes API and enforces access via token review and subject access review. Demo archives Tekton PipelineRuns and batch Jobs by label and name pattern, is pre‑alpha with a Helm chart and CNCF sandbox intent.
AI-written summary · about
chapters (18)
- 0:00KubeArchive introduction
- 1:29Why Kubernetes shouldn’t store state
- 3:14Small objects filling etcd
- 5:22Tecton Results limitations
- 6:03KubeArchive goals
- 7:52How KubeArchive works
- 9:59Object workflow in KubeArchive
- 10:42REST API for archived data
- 12:34KubeArchive custom resources
- 14:25Live demo setup
- 17:00Archiving pipeline runs
- 20:11Archiving Kubernetes Jobs
- 23:07Retrieving data with service accounts
- 28:46Namespace-scoped archived access
- 29:50Permission enforcement in KubeArchive
- 30:53Current features and roadmap
- 33:01How to learn more
- 34:26Q&A
related talks
36mHow To Build Kube-Native CI/CD Pipelines for Containerized Databases
57mSecure Kubernetes data management and scaling with Kasten K10
18mWhen Kubernetes Becomes an Interface, Not a Cluster
01h 00mKubernetes and cloud-native data management
01h 40mInside Kubernetes Architecture Fundamentals
45mFrom Laptop to Cloud: Developing Cloud Native Applications with Containerized Databases
27mApplication Aware Backups With Kanister And Kopia
26mChallenges of Managing Stateful Applications in Kubernetes