Using Kubernetes to Change Legacy Systems and Processes in the Public Sector
summary
Kubernetes becomes the tool to modernize a decades-old public sector stack: two clusters (production + not‑production) host apps isolated by namespaces and enforced by network policies, access uses OpenID Connect against Azure AD, and monitoring runs on Prometheus metrics with generic dashboards. A Go-based "nice deployment daemon" and Landscaper automate app rollout; Rook provides storage and a Postgres operator replaces Oracle over time. Results: phased migrations, fewer test environments, developers gain controlled production access, and early production apps run successfully.
AI-written summary · about
chapters (31)
- 0:00Introduction
- 0:51Problems
- 1:42Overview
- 2:03About me
- 3:10Migration to Kubernetes
- 4:26Norways IT Department
- 5:21History of the IT Department
- 7:41Private Cloud
- 8:34Culture
- 9:21Vernon Histor
- 11:25The Problems
- 11:52Data Centers
- 13:09Kubernetes
- 14:18User Growth
- 15:24Test Environments
- 16:52Solution
- 18:20Access Control
- 19:21OpenID Connect
- 20:37Network Process
- 22:40Monitoring
- 22:56Monitoring Solutions
- 25:08Creating New Apps
- 25:40Nice Deployment
- 28:22Bad Batch Jobs
- 29:12Batch Jobs
- 29:52Architecture
- 30:25Continuous Delivery
- 30:58Creating the Platform
- 31:46Migrating for Storage
- 32:21Open Source
- 34:35DevOps Culture
related talks
35mModernize Legacy Systems with Kubernetes
29mA history of GOV.UK's infrastructure: the long voyage to Kubernetes
44mMigrating a Monolith to Kubernetes
35mKubernetes journey
20mApplication Modernization: From Concept to Kubernetes
16mModernizing a legacy app using Windows Containers and Kubernetes: Challenges and Lessons learned
46mAround & After Kubernetes: The Principles and Ideas that Guide Us
12mHow to Effectively Manage Kubernetes in a Regulated Environment