Observability Overview
Observability is essential for understanding the behavior, performance, and health of your integrations in production. WSO2 Integrator provides built-in support for the three pillars of observability: metrics, logging, and distributed tracing. Choose from WSO2-managed solutions, self-managed open-source stacks, or commercial platforms based on your deployment model and requirements.
The Three Pillars of Observability
| Pillar | Purpose | Key Metrics | Built-in Support |
|---|---|---|---|
| Metrics | Quantitative measurements of system behavior | Request counts, latency, error rates, throughput | Prometheus-compatible endpoint |
| Logging | Structured event records for debugging and auditing | Log entries with context, error details, request tracing | Ballerina log module with configurable levels |
| Distributed Tracing | End-to-end request flow across services | Span duration, service dependencies, bottlenecks | OpenTelemetry-based tracing |
Architecture
WSO2 Provided Solutions
WSO2 provides fully managed observability solutions for integrations deployed on the WSO2 platform.
| Solution | Best For | Features | Setup Complexity |
|---|---|---|---|
| Integration Control Plane (ICP) | On-premise & hybrid deployments | Service inventory, real-time monitoring, log aggregation, deployment tracking | Low |
| WSO2 Integration Platform | Cloud-native integrations | Built-in dashboards, alerting, live logs, distributed tracing, diagnostics | Very Low |
| Moesif | API analytics & monitoring | API usage tracking, request/response inspection, usage-based billing, alerting | Very Low |
When to choose WSO2 solutions:
- You want zero or minimal setup for observability
- You need built-in integration with WSO2 platform
- You require compliance with WSO2 enterprise support
- For API analytics (Moesif), you need deep insight into API usage patterns and customer behavior
Self-Managed Solutions (Open Source)
Deploy and manage your own observability stack. Ideal for organizations with existing infrastructure investments or specific compliance requirements.
Metrics (Prometheus)
- Metrics Overview – Enable Prometheus metrics, configure scraping, and define custom metrics
When to use: Open-source, self-hosted metrics collection from your Ballerina integrations.
Logging
- Logging Overview – Configure structured logging and log aggregation approaches
When to use: Full-text log search, complex log processing, multi-service log correlation.
Distributed Tracing (Jaeger or Zipkin)
- Jaeger Setup – Production-grade distributed tracing
- Zipkin Setup – Lightweight distributed tracing alternative
When to use: Trace requests across service boundaries, identify latency bottlenecks, debug request failures.
Commercial Managed Solutions
Deploy your integration on managed cloud platforms with built-in observability. Lower operational overhead, SLA guarantees, and dedicated support.
| Platform | Best For | Metrics | Logging | Tracing | Setup |
|---|---|---|---|---|---|
| New Relic | Multi-cloud environments | ✅ Via OTLP | ✅ Via forwarder | ✅ Via OTLP | Low |
Recipes: End-to-End Solutions
Choose a recipe based on your deployment scenario and infrastructure.
Recipe 1: Datadog Full-Stack Observability
Tech Stack: Datadog Agent + Datadog Cloud
Managed cloud observability with minimal setup.
- Install Datadog Agent as sidecar
- Automatic metric, log, and trace collection
- Datadog dashboards and monitors
- Native APM and service maps