Skip to content

observability-monitoring.md

Latest commit

 

History

History
132 lines (83 loc) · 6.08 KB

observability-monitoring.md

File metadata and controls

132 lines (83 loc) · 6.08 KB
Raw

Observability & Monitoring

Table of Contents

Quick Start

Pick a repo below, copy the Step 1 prompt into a new Devin session, and let Devin add structured logging, metrics, and tracing to the application. No prerequisites beyond repo access.

Challenge

Add observability instrumentation to an application — structured logging, metrics collection, distributed tracing, and health dashboards. This exercises Devin's ability to integrate observability libraries and configure monitoring infrastructure.

Target Outcomes

  • Structured JSON logging with correlation IDs
  • Prometheus-compatible metrics endpoint
  • Health check and readiness endpoints
  • OpenTelemetry tracing (or equivalent) integrated
  • PR with observability improvements

What Participants Will Learn

  • How Devin integrates observability libraries into existing applications
  • Structured logging vs. unstructured logging
  • Metrics, traces, and logs — the three pillars of observability
  • How to configure monitoring endpoints for production readiness

Devin Features Exercised

  • Library integration (Winston, Micrometer, OpenTelemetry)
  • Configuration management
  • Multi-file changes across application layers
  • PR creation

Difficulty

Intermediate to Advanced

Estimated Time

60 minutes

Repositories

timesheet-app

Repository: timesheet-app

Node.js/Express application — add Winston structured logging, Prometheus metrics, and OpenTelemetry tracing.

Step 1: Paste into Devin

Add observability to timesheet-app: (1) Replace console.log with Winston structured JSON logging with correlation IDs per request, (2) Add a /metrics endpoint using prom-client for Prometheus scraping (request count, latency histogram, error rate), (3) Add OpenTelemetry auto-instrumentation for HTTP and Express.

Step 2: Research with Ask Devin

  • "What logging patterns does timesheet-app currently use? How inconsistent is the logging?"
  • "What metrics are most valuable for a CRUD application — request rate, error rate, latency, or database query times?"

Step 3 (Optional): Read the DeepWiki

Open the repo's DeepWiki page to understand the middleware chain and request lifecycle. Plan where to inject logging, metrics, and tracing middleware.

Step 4 (Optional): Review & Give Feedback

  • Review the diff — are correlation IDs propagated through the entire request lifecycle?
  • Leave a comment asking Devin to add a Docker Compose file with Prometheus and Grafana for local observability

uc-spring-boot-upgrade-microservice-extraction

Repository: uc-spring-boot-upgrade-microservice-extraction

Spring Boot application — add Spring Boot Actuator, Micrometer metrics, and structured logging with Logback.

Step 1: Paste into Devin

Add observability to uc-spring-boot-upgrade-microservice-extraction: (1) Enable Spring Boot Actuator with health, info, metrics, and prometheus endpoints, (2) Add Micrometer Prometheus registry for metrics export, (3) Configure Logback for structured JSON logging with MDC correlation IDs, (4) Add custom metrics for article and user operations.

Step 2: Research with Ask Devin

  • "Which Spring Boot Actuator endpoints should be exposed vs. secured in production?"
  • "What custom business metrics would be valuable — article creation rate, user registration rate, API error rate?"

Step 3 (Optional): Read the DeepWiki

Open the repo's DeepWiki page to understand the application layers and identify where to add custom metrics. Map out the request flow for MDC correlation ID propagation.

Step 4 (Optional): Review & Give Feedback

  • Review the diff — are Actuator endpoints properly secured? Are custom metrics named following Micrometer conventions?
  • Leave a comment asking Devin to add a Grafana dashboard JSON that visualizes the key metrics

Key Takeaways

  • Observability instrumentation is a cross-cutting concern that touches many files — Devin handles the multi-file coordination (middleware, configuration, dependencies) in a single session
  • Adding structured logging, metrics, and tracing is well-defined work that benefits from Devin's ability to follow library documentation and integrate patterns consistently
  • The /metrics and /health endpoints Devin generates are the foundation for production monitoring — they connect directly to Prometheus, Grafana, and alerting pipelines

Going Further

Observability connects directly to incident-response automation (see When to Use Devin → Event-Driven):

  • Alert-triggered investigation — Once metrics endpoints are in place, connect alerting rules (Prometheus AlertManager, Datadog monitors) to the Devin API. When an alert fires (e.g., error rate > 5%), Devin queries logs and traces via MCP, identifies the root cause, and opens a fix PR — often before the on-call engineer is paged
  • Dashboard-as-code maintenance — Schedule a recurring Devin session that compares Grafana dashboard definitions against the application's current metrics endpoints, adding panels for new metrics and removing panels for deprecated ones
  • Observability coverage audits — A periodic session analyzes the codebase for unstructured logging (e.g., console.log, System.out.println) and opens PRs to migrate remaining instances to the structured logging framework