Running a real-world mission-critical system on Azure
summary
A mission-critical .NET Core system is reimplemented as containers on Azure Kubernetes Service with CI/CD via Visual Studio 2019 → Azure DevOps, Helm charts and Azure Container Registry, plus ARM templates for infrastructure as code. Production deployment uses controlled rollouts, automatic healing and a Slack bot for chatops, telemetry with App Insights, incident runbooks and a public status page. Resilience uses retry + circuit breaker + outbox + leader election patterns and a daily automated failover to validate redundancy.
AI-written summary · about
chapters (29)
- 0:00Intro
- 0:15What is a missioncritical system
- 1:40The gas plant
- 2:08The old system
- 4:16What is DevOps
- 4:27Micro outages
- 5:31Enterprise architecture
- 6:46Functional Cohesion
- 7:28User Interface
- 9:01Feedback loops
- 9:20Errors
- 9:57Outgoing requests
- 11:10Customer Interaction
- 11:59Monitoring
- 14:09Continuous Delivery
- 16:12Deploying Kubernetes
- 18:07Automation
- 20:03Failover
- 20:45Death Star
- 21:20Microservices
- 22:15Smaller version
- 23:15Design patterns
- 23:41Groundhog Day
- 24:50Circuit Breaker Pattern
- 26:54Outbox Pattern
- 28:18Leader Election Pattern
- 30:41Demo Telemetry
- 33:56Call the Shots
- 36:34Questions
related talks
35mRunning a DevOps Style Production Docker Cluster Using the Microsoft Platform
59mRunning a real world, mission-critical system on Azure
52mBring your ASP.NET Core solutions to Kubernetes in Azure
46mReal World DevOps
49mModern CI/CD Pipeline with Azure DevOps
01h 00mPractical Container Scenarios in Azure
40mDevOps and Kubernetes in VSTS on Azure and Beyond
01h 00mBring your ASP.NET Core solutions to Kubernetes in Azure