--- name: langfuse-common-errors-9b857528d9d5 description: | Diagnose and fix common Langfuse errors and exceptions. Use when encountering Langfuse errors, debugging missing traces, or troubleshooting integration issues. Trigger with phrases like "langfuse error", "fix langfuse", "langfuse not working", "debug langfuse", "traces not appearing". allowed-tools: Read, Grep, Bash(curl:*) version: 1.0.0 license: MIT author: Jeremy Longshore --- # Langfuse Common Errors ## Overview Quick reference for the top 10 most common Langfuse errors and their solutions. ## Prerequisites - Langfuse SDK installed - API credentials configured - Access to error logs ## Instructions ### Step 1: Identify the Error Check error message and code in your logs or console. ### Step 2: Find Matching Error Below Match your error to one of the documented cases. ### Step 3: Apply Solution Follow the solution steps for your specific error. ## Output - Identified error cause - Applied fix - Verified resolution ## Error Handling ### 1. Authentication Failed **Error Message:** ``` Langfuse: Unauthorized - Invalid API key Error: 401 Unauthorized ``` **Cause:** API key is missing, expired, or mismatched with host. **Solution:** ```bash # Verify environment variables are set echo "Public: $LANGFUSE_PUBLIC_KEY" echo "Secret: ${LANGFUSE_SECRET_KEY:0:10}..." # Partial for security echo "Host: $LANGFUSE_HOST" # Test connection curl -X GET "$LANGFUSE_HOST/api/public/health" \ -H "Authorization: Basic $(echo -n "$LANGFUSE_PUBLIC_KEY:$LANGFUSE_SECRET_KEY" | base64)" ``` --- ### 2. Traces Not Appearing in Dashboard **Symptom:** Code runs without errors but traces don't show up. **Cause:** Data not flushed before process exits. **Solution:** ```typescript // Always flush before exit await langfuse.flushAsync(); // Or use shutdown for clean exit await langfuse.shutdownAsync(); // Register shutdown handler process.on("beforeExit", async () => { await langfuse.shutdownAsync(); }); ``` --- ### 3. Network/Connection Errors **Error Message:** ``` FetchError: request to https://cloud.langfuse.com failed ECONNREFUSED / ETIMEDOUT ``` **Cause:** Network connectivity or firewall issues. **Solution:** ```bash # Test connectivity curl -v https://cloud.langfuse.com/api/public/health # Check DNS resolution nslookup cloud.langfuse.com # For self-hosted, verify host URL curl -v $LANGFUSE_HOST/api/public/health ``` ```typescript // Increase timeout if needed const langfuse = new Langfuse({ requestTimeout: 30000, // 30 seconds }); ``` --- ### 4. Missing Token Usage **Symptom:** Generations appear but token counts are 0. **Cause:** Usage data not captured from LLM response. **Solution:** ```typescript // For streaming, enable usage in options const stream = await openai.chat.completions.create({ model: "gpt-4", messages, stream: true, stream_options: { include_usage: true }, // Important! }); // Manually add usage if not available generation.end({ output: content, usage: { promptTokens: response.usage?.prompt_tokens || 0, completionTokens: response.usage?.completion_tokens || 0, }, }); ``` --- ### 5. Trace/Span Not Ending **Symptom:** Traces show as "in progress" indefinitely. **Cause:** Missing `.end()` call on spans/generations. **Solution:** ```typescript // Always end in try/finally const span = trace.span({ name: "operation" }); try { const result = await doWork(); span.end({ output: result }); return result; } catch (error) { span.end({ level: "ERROR", statusMessage: String(error) }); throw error; } // Or use finally const span = trace.span({ name: "operation" }); try { return await doWork(); } finally { span.end(); } ``` --- ### 6. Duplicate Traces **Symptom:** Same operation creates multiple traces. **Cause:** Client instantiated multiple times or missing singleton. **Solution:** ```typescript // Use singleton pattern let langfuseInstance: Langfuse | null = null; export function getLangfuse(): Langfuse { if (!langfuseInstance) { langfuseInstance = new Langfuse(); } return langfuseInstance; } // Don't create new instance per request // BAD: const langfuse = new Langfuse(); // in handler // GOOD: const langfuse = getLangfuse(); ``` --- ### 7. SDK Version Mismatch **Error Message:** ``` TypeError: langfuse.trace is not a function Property 'observeOpenAI' does not exist ``` **Cause:** Outdated SDK or breaking API change. **Solution:** ```bash # Check current version npm list langfuse # Update to latest npm install langfuse@latest # Check changelog for breaking changes # https://github.com/langfuse/langfuse-js/releases ``` --- ### 8. Environment Variable Not Loaded **Error Message:** ``` Langfuse: Missing required configuration - publicKey ``` **Cause:** .env file not loaded or wrong variable names. **Solution:** ```bash # Verify .env file exists and has correct names cat .env | grep LANGFUSE # Required variable names: # LANGFUSE_PUBLIC_KEY=pk-lf-... # LANGFUSE_SECRET_KEY=sk-lf-... # LANGFUSE_HOST=https://cloud.langfuse.com ``` ```typescript // Load .env in your app import "dotenv/config"; // At top of entry file // Or specify env file import { config } from "dotenv"; config({ path: ".env.local" }); ``` --- ### 9. Self-Hosted Connection Issues **Error Message:** ``` Failed to connect to localhost:3000 Certificate verification failed ``` **Cause:** Self-hosted instance not running or SSL issues. **Solution:** ```bash # Check if Langfuse is running docker ps | grep langfuse curl http://localhost:3000/api/public/health # For HTTPS without valid cert export NODE_TLS_REJECT_UNAUTHORIZED=0 # Development only! ``` ```typescript // Verify host URL doesn't have trailing slash const langfuse = new Langfuse({ baseUrl: "http://localhost:3000", // No trailing slash }); ``` --- ### 10. Rate Limiting **Error Message:** ``` Error: 429 Too Many Requests Retry-After: 60 ``` **Cause:** Exceeded Langfuse API rate limits. **Solution:** ```typescript // Increase batch size to reduce requests const langfuse = new Langfuse({ flushAt: 50, // Batch more events flushInterval: 30000, // Flush less often }); // See langfuse-rate-limits skill for advanced handling ``` ## Quick Diagnostic Commands ```bash # Check Langfuse cloud status curl -s https://status.langfuse.com # Verify API connectivity curl -I https://cloud.langfuse.com/api/public/health # Check local environment env | grep LANGFUSE # Test auth curl -X GET "https://cloud.langfuse.com/api/public/traces" \ -H "Authorization: Basic $(echo -n "$LANGFUSE_PUBLIC_KEY:$LANGFUSE_SECRET_KEY" | base64)" \ | head -c 200 ``` ## Escalation Path 1. Collect evidence with `langfuse-debug-bundle` 2. Check [Langfuse Status](https://status.langfuse.com) 3. Search [GitHub Issues](https://github.com/langfuse/langfuse/issues) 4. Join [Discord Community](https://langfuse.com/discord) 5. Contact support with debug bundle ## Resources - [Langfuse Troubleshooting](https://langfuse.com/docs/get-started) - [Langfuse GitHub Issues](https://github.com/langfuse/langfuse/issues) - [Langfuse Discord](https://langfuse.com/discord) ## Next Steps For comprehensive debugging, see `langfuse-debug-bundle`. --- Source attribution: Dicklesworthstone Original project: https://github.com/Dicklesworthstone/pi_agent_rust/blob/f28fc46ee0db8e710d95b8b55592cd38d60f2b26/tests/ext_conformance/artifacts/plugins-community/plugins/saas-packs/langfuse-pack/skills/langfuse-common-errors/SKILL.md License: MIT License information: https://opensource.org/license/mit Original upstream instructions; downloaded from Mega AI Hub. ## Upstream license notice MIT License (with OpenAI/Anthropic Rider) Copyright (c) 2026 Jeffrey Emanuel Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: ADDITIONAL RIDER / RESTRICTION (OpenAI / Anthropic): This rider is part of the "conditions" of this License. In the event of any conflict between this rider and any other portion of this License, this rider controls. "Restricted Parties" means OpenAI, L.L.C.; Anthropic, PBC; any of their respective Affiliates; and any person or entity acting directly or indirectly on behalf of, for the benefit of, or under the direction of any of the foregoing (including any officer, director, employee, contractor, agent, consultant, service provider, or representative). Notwithstanding any other provision of this License, no rights are granted to any Restricted Party. Any purported license, sublicense, assignment, transfer, or other permission to any Restricted Party is null and void absent the express prior written permission of Jeffrey Emanuel. You may not provide, disclose, distribute, sublicense, sell, lease, lend, host, make available, or otherwise permit access to the Software or any derivative work of the Software (as defined in applicable copyright law) (collectively, "Derivative Works") to or for any Restricted Party. For purposes of this rider, "use" includes, without limitation: copying, modifying, merging, publishing, distributing, sublicensing, selling, transferring, making available, hosting, deploying, executing, benchmarking, testing, analyzing, indexing, or incorporating the Software or any Derivative Works into any dataset, training corpus, evaluation harness, or pipeline for machine learning or other automated systems. This rider applies to the Software and all Derivative Works. As a condition of use, you agree that this rider is a precondition to exercising any rights under this License, and you agree that any distribution of the Software or any Derivative Works must include this rider provision unmodified. Any breach of this rider automatically and immediately terminates the permissions granted by this License. Upon termination, you must immediately cease all use and distribution of the Software and any Derivative Works and destroy all copies under your control. You agree that a breach of this rider would cause irreparable harm and that Jeffrey Emanuel may seek injunctive or other equitable relief to enforce this rider, in addition to any other remedies available at law. To the maximum extent permitted by applicable law, the prevailing party in any action to enforce this rider shall be entitled to recover reasonable attorneys' fees and costs. For purposes of this rider, "Affiliate" means any entity that directly or indirectly controls, is controlled by, or is under common control with the specified party. "Control" means ownership of more than 50% of the voting securities or other ownership interest, or the power to direct management or policies by contract or otherwise. The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. # Mega AI Hub resource metadata Creator / publisher: Dicklesworthstone Original project: https://github.com/Dicklesworthstone/pi_agent_rust/blob/f28fc46ee0db8e710d95b8b55592cd38d60f2b26/tests/ext_conformance/artifacts/plugins-community/plugins/saas-packs/langfuse-pack/skills/langfuse-common-errors/SKILL.md Attribution basis: Source repository record Fields of work: Development & Engineering Roles: Backend Developer, Data Engineer, ML / AI Engineer Specialties: API, Backend Tools: Langfuse Country applicability: Country not specified Classification basis: Source catalog labels Instruction SHA-256: 55331de5af83b2a9685dde19ad9f7bb71069232320d9a37a086dc18880ed1315