Langfuse Production Checklist
Contents
Overview
Comprehensive checklist for deploying Langfuse observability to production with verified configuration, error handling, graceful shutdown, and monitoring.
Prerequisites
- Completed development and staging testing
- Access to production secrets management
- Production Langfuse project created
Instructions
Step 1: Verify Authentication & Security
Confirm production API keys are created, stored in a secrets manager, NOT in code/git, and isolated per environment. Verify PII scrubbing is enabled.
Step 2: Configure SDK Settings
Set flushAt (15-50), flushInterval (5-10s), requestTimeout (10-30s), enabled (true for prod). Use singleton client pattern.
Step 3: Implement Production Error Handling
Wrap all trace operations in try/catch. Call .end() in finally blocks. Log errors to Langfuse with ERROR level. Register shutdownAsync() on SIGTERM/SIGINT.
Step 4: Run Pre-Deployment Verification
Validate env vars are set, API keys are production keys (prefix pk-lf-), trace creation works, and graceful shutdown completes.
Step 5: Configure Production Monitoring
Expose metrics endpoint tracking traces created/failed, flush latency, and error rate.
See detailed implementation for advanced patterns.
Output
- Verified production configuration
- Error handling with graceful degradation
- Graceful shutdown on process signals
- Metrics endpoint for monitoring
- All checklist items validated
Error Handling
| Issue |
Cause |
Solution |
| Missing traces in prod |
Flush not called |
Verify shutdown handler |
| High latency |
Large batches |
Reduce flushAt |
| Memory growth |
Client recreation |
Use singleton pattern |
| Lost traces on deploy |
No graceful shutdown |
Add SIGTERM handler |
Examples
Production Checklist Summary
| Category |
Key Items |
| Auth & Security |
Prod keys, secrets manager, env isolation, PII scrubbing |
| SDK Config |
flushAt=25, flushInterval=5000, singleton client |
| Error Handling |
try/catch, .end() in finally, ERROR level traces |
| Performance |
Async tracing, batching, no memory leaks |
| Observability |
Trace URL logging, SDK error monitoring, alerts |
Recommended Production Config
{
flushAt: 25,
flushInterval: 5000, # 5000: 5 seconds in ms
requestTimeout: 15000, # 15000 = configured value
enabled: process.env.NODE_ENV === "production",
debug: false,
}
Resources
1---2name: langfuse-prod-checklist3description: |4 Langfuse production readiness checklist and verification.5 Use when preparing to deploy Langfuse to production,6 validating production configuration, or auditing existing setup.7 Trigger with phrases like "langfuse production", "langfuse prod ready",8 "deploy langfuse", "langfuse checklist", "langfuse go live".9allowed-tools: Read, Write, Edit, Bash(npm:*), Grep10version: 1.0.011license: MIT12author: Jeremy Longshore <jeremy@intentsolutions.io>13compatible-with: claude-code, codex, openclaw14metadata:15 source: https://skillmd.com/skills/micsapp/langfuse-prod-checklist16---1718# Langfuse Production Checklist1920## Contents21- [Overview](#overview)22- [Prerequisites](#prerequisites)23- [Instructions](#instructions)24- [Output](#output)25- [Error Handling](#error-handling)26- [Examples](#examples)27- [Resources](#resources)2829## Overview30Comprehensive checklist for deploying Langfuse observability to production with verified configuration, error handling, graceful shutdown, and monitoring.3132## Prerequisites33- Completed development and staging testing34- Access to production secrets management35- Production Langfuse project created3637## Instructions3839### Step 1: Verify Authentication & Security40Confirm production API keys are created, stored in a secrets manager, NOT in code/git, and isolated per environment. Verify PII scrubbing is enabled.4142### Step 2: Configure SDK Settings43Set `flushAt` (15-50), `flushInterval` (5-10s), `requestTimeout` (10-30s), `enabled` (true for prod). Use singleton client pattern.4445### Step 3: Implement Production Error Handling46Wrap all trace operations in try/catch. Call `.end()` in `finally` blocks. Log errors to Langfuse with ERROR level. Register `shutdownAsync()` on SIGTERM/SIGINT.4748### Step 4: Run Pre-Deployment Verification49Validate env vars are set, API keys are production keys (prefix `pk-lf-`), trace creation works, and graceful shutdown completes.5051### Step 5: Configure Production Monitoring52Expose metrics endpoint tracking traces created/failed, flush latency, and error rate.5354See [detailed implementation](${CLAUDE_SKILL_DIR}/references/implementation.md) for advanced patterns.5556## Output57- Verified production configuration58- Error handling with graceful degradation59- Graceful shutdown on process signals60- Metrics endpoint for monitoring61- All checklist items validated6263## Error Handling64| Issue | Cause | Solution |65|-------|-------|----------|66| Missing traces in prod | Flush not called | Verify shutdown handler |67| High latency | Large batches | Reduce `flushAt` |68| Memory growth | Client recreation | Use singleton pattern |69| Lost traces on deploy | No graceful shutdown | Add SIGTERM handler |7071## Examples7273### Production Checklist Summary74| Category | Key Items |75|----------|-----------|76| Auth & Security | Prod keys, secrets manager, env isolation, PII scrubbing |77| SDK Config | flushAt=25, flushInterval=5000, singleton client |78| Error Handling | try/catch, .end() in finally, ERROR level traces |79| Performance | Async tracing, batching, no memory leaks |80| Observability | Trace URL logging, SDK error monitoring, alerts |8182### Recommended Production Config83```typescript84{85 flushAt: 25,86 flushInterval: 5000, # 5000: 5 seconds in ms87 requestTimeout: 15000, # 15000 = configured value88 enabled: process.env.NODE_ENV === "production",89 debug: false,90}91```9293## Resources94- [Langfuse Production Guide](https://langfuse.com/docs/deployment)95- [Langfuse Self-Hosting](https://langfuse.com/docs/deployment/self-host)96- [Langfuse SDK Reference](https://langfuse.com/docs/sdk)