Running Dapr in Production
summary
Concrete ops checklist for running Dapr in production: deploy a highly available control plane with three replicas in the dapr-system namespace on a Kubernetes cluster with at least three worker nodes; install/manage with the Dapr CLI, Helm or GitHub Actions; perform zero-downtime upgrades (demoed 1.4.0 → 1.4.2) by upgrading control plane then rolling sidecars; use sidecars with mutual TLS via Sentry (default cert TTL 24h, rotation configurable), configure JWT app↔Dapr tokens, use scoped components and external secret stores, and enable Zipkin tracing and Prometheus metrics (port 9090).
AI-written summary · about
chapters (8)
- 0:00Start
- 0:34Microservice Development Dapr Goals
- 3:58Production Recommendations Deploying Dapr
- 8:22Upgrading Dapr Zero Downtime
- 19:30Production Recommendations Security: Namespaces
- 20:50Production Recommendations Security: Secrets
- 23:14Secrets Component YAMLS and Secrets
- 26:17Production Recommendations Tracing and Metrics
related talks
49mDapr in Practice
31mBuilding scalable distributed systems powered by Dapr OSS.
51mDay 2 Session 2
45mDapr Workflows: How to Write Reliable, Fault-Tolerant Applications...
25mSleep Better at Night: Dapr's Approach to Secure Cloud Native Development
56mBuilding Microservices With Dapr and .NET
04mProject Lightning Talk: Dapr: Start Building Distributed Applications with Ease Using Building Block APIs
51mDapr for building distributed .NET Core applications