The chat model worked fine at first, but it stopped working after the upgrade.

[oilycn]

Describe the problem/error/question

The chat model node does not show the running status, but it can output the results at the end. The output of the chat model is not shown in the log of the ai agent, and it takes a long time to reply.
The background log shows:
Error in handler N8nLlmTracing, handleLLMStart: TypeError: fetch failed
Error in handler N8nLlmTracing, handleLLMEnd: TypeError: fetch failed

What is the error message (if any)?

There is no error message, but the output of the chat model is not displayed in the ai agent log

Please share your workflow/screenshots/recording

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Debug info

core

• n8nVersion: 1.75.0
• platform: docker (self-hosted)
• nodeJsVersion: 20.18.0
• database: sqlite
• executionMode: regular
• concurrency: -1
• license: enterprise (production)

storage

• success: all
• error: all
• progress: false
• manual: true
• binaryMode: memory

pruning

• enabled: true
• maxAge: 336 hours
• maxCount: 10000 executions

client

• userAgent: mozilla/5.0 (windows nt 10.0; win64; x64) applewebkit/537.36 (khtml, like gecko) chrome/131.0.0.0 safari/537.36 edg/131.0.0.0
• isTouchDevice: false

security

• secureCookie: false

Generated at: 2025-01-17T12:25:28.764Z}

[Joffcom]

Hey @oilycn,

We have created an internal ticket to look into this which we will be tracking as "N8N-8140"

[burivuhster]

Hey @oilycn , can you please help us to reproduce the issue by providing the workflow JSON?

[oilycn]

__.json @burivuhster

[jeanpaul]

@oilycn It looks like you used a different model provider by setting the BaseURL in your OpenAI credentials. Using the gemini model that you also had present in your workflow did seem to work properly. Can you verify that the URL in your OpenAI credentials work correctly?

[oilycn]

yes,the URL is worded in my OpenAI credentials.The application is functioning correctly on a VPS in Hong Kong, but not on a server located in Mainland China. During local testing, the base URL works fine, and it also worked fine in previous versions. The issue seems to have appeared in one or two releases after the base URL was changed to a configuration key interface.

[oilycn]

@jeanpaul

[jeanpaul]

Hi @oilycn -- can you verify that this also breaks with other URLs that follow the OpenAI API spec? I'm unable to help you with this specific host, but I've verified that other hosts do work properly...