Keeping an eye on your serverless containers
summary
A demo builds a retirement-home site called cuddle cube that registers server containers and runs on AWS using Fargate, App Mesh, DynamoDB, CloudWatch and X‑Ray. Envoy sidecars emit telemetry so distributed tracing reveals a 15‑second DynamoDB timeout as the failure point. Logs and CloudWatch metrics fill in CPU and errors. Setup requires minimal code changes for HTTP context propagation and some instrumentation; the approach works but setup and integrations are finicky.
AI-written summary · about
chapters (7)
- 0:00Introduction and setup
- 3:38Introducing Cuddle Cube concept
- 10:47Defining observability pillars
- 17:12Service mesh for serverless containers
- 24:22Live demo of Cuddle Cube
- 31:12Challenges and limitations
- 33:19Key takeaways and conclusion
related talks
50mBuild, Deploy, Debug and Tune Serverless Microservices
41mLive Coding on Observability for Serverless
44mServerless and containers are coming together: an AWS perspective
47mServerless Compute at the Heart of Your EDA
51mServerless Compute at the Heart of Your EDA
43mBringing EDA to the Containers World
49mMonitoring serverless applications on AWS: the monitoring monster becomes your friend.
47mBuild your next microservices application with modern AWS services