Overview
Request Logs support a wide range of use cases. Each guide below walks through a specific scenario with step-by-step instructions, real examples, and actionable advice.
Debugging & troubleshooting
Debugging Production Issues
When a user reports that an AI feature isn't working correctly, use Request Logs to trace the complete request lifecycle. Filter to live production traffic, inspect the timeline for failover attempts and webhook failures, and examine request/response payloads to identify the root cause.
Troubleshooting Failures
Systematically diagnose and resolve request failures. Categorize errors by type (provider failures, webhook delivery issues, callback timeouts), apply targeted fixes, and verify resolution through follow-up testing.
Testing & validation
Testing Workflows
Validate workflow changes safely before deploying to production. Use Playground (Test mode) for free, creditless testing, Playground (Production) for real provider validation, and the test mode API for CI/CD integration.
Monitoring & analysis
Cost Analysis
Understand AI spending patterns by tracking token usage and estimated pricing per request. Identify high-cost requests, compare provider costs, and apply optimization strategies like prompt trimming and model-task matching.
Performance Monitoring
Monitor request latency, establish baseline metrics, and identify performance bottlenecks. Compare provider speeds, track peak-hour degradation, and optimize for the best user experience.
Provider Reliability Analysis
Analyze provider failure rates, track failover patterns, and use historical data to make informed decisions about workflow provider priority. Measure outage impact and set up reliability alerts.
Webhook Delivery Monitoring
Ensure webhooks reach your backend reliably. Monitor delivery status, diagnose common issues (timeouts, connection errors, SSL problems), use retry functionality, and track callback completion for event-driven workflows.
Quick reference
| Scenario | Recommended page |
|---|---|
| User reports AI isn't working | Debugging Production Issues |
| Requests are failing | Troubleshooting Failures |
| Validate before deploying | Testing Workflows |
| AI spending is too high | Cost Analysis |
| Responses feel slow | Performance Monitoring |
| Provider outages | Provider Reliability |
| Backend not receiving webhooks | Webhook Monitoring |
Next steps
- Best Practices — General guidelines for Request Logs
- Timeline Components — Deep dive into timeline details
- Back to Observability — Return to the overview