n8n - 💡(How to fix) Fix Code node hangs with AI Agent HTTP Request Tool present [1 comments, 2 participants]

Task execution timed out after 300 seconds

The task runner was taking too long on this task, so it was suspected of being unresponsive and restarted, and the task was aborted.

Stack trace:
Error: Task execution timed out after 300 seconds
    at TaskBroker.handleTaskTimeout (/usr/local/lib/node_modules/n8n/src/task-runners/task-broker/task-broker.service.ts:504:4)
    at Timeout.<anonymous> (/usr/local/lib/node_modules/n8n/src/task-runners/task-broker/task-broker.service.ts:476:15)
    at listOnTimeout (node:internal/timers:605:17)
    at processTimers (node:internal/timers:541:7)

Actual behavior
The Code node times out after 300 seconds.

Execution details:

Status: Error
Duration: 5m 0.131s
Failed node: Product Router
n8n version: 2.19.5 Self Hosted
The input item is small, for example:

{
  "data": {
    "inbox_id": 1,
    "inbox_title": "development",
    "conv_id": 385,
    "message_id": 6549,
    "message_content": "<p>Hello</p>",
    "message_type": "incoming",
    "account_id": 2,
    "content_type": "text",
    "event": "message_created"
  }
}
Expected behavior
The Code node should finish immediately. The presence of an AI Agent / HTTP Request Tool elsewhere in the workflow should not cause unrelated Code nodes to hang or hit the task runner timeout.

Workflow structure
Simplified structure:

Webhook
-> Config (Code)
-> Set / Normalize webhook data
-> Product Router (Code)  <-- times out here
-> IF
-> Load conversation history
-> Process history (Code)
-> AI Agent
   ├─ Chat Model
   └─ HTTP Request Tool for knowledge base / RAG
-> downstream nodes
The HTTP Request Tool is used as a knowledge-base / RAG tool. It cannot simply be replaced by a single main-flow HTTP Request node, because the AI Agent needs to decide dynamically when and how many times to query the knowledge base.

### To Reproduce

The issue seems tied to the combination of:

Self-hosted n8n
Task runners enabled
Code node
AI Agent node
HTTP Request Tool connected to or present for the AI Agent
Removing the AI Tool or avoiding Code nodes appears to avoid the timeout.

Environment
n8n version: 2.19.5
Deployment: Self Hosted
Task runners: enabled
Execution type: webhook workflow
Why this seems like an n8n runtime/task-runner issue
The failing Code node contains no async work, no external HTTP calls, and no loop. The stack trace points to TaskBroker.handleTaskTimeout, and the failure happens exactly at the default task runner timeout of 300 seconds.

This suggests the task runner or workflow initialization/AI Tool handling is hanging or blocking execution, rather than the Code node itself taking 300 seconds.

### Expected behavior

The Code node should finish immediately. The presence of an AI Agent / HTTP Request Tool elsewhere in the workflow should not cause unrelated Code nodes to hang or hit the task runner timeout.

### Debug Info

# Debug info

## core

- n8nVersion: 2.19.5
- platform: docker (self-hosted)
- nodeJsVersion: 24.14.1
- nodeEnv: production
- database: postgres
- executionMode: regular
- concurrency: -1
- license: enterprise (production)
- consumerId: c56f5a81-3ec7-461b-949e-175e357cd4ab

## storage

- success: all
- error: all
- progress: false
- manual: true
- binaryMode: filesystem

## pruning

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

## client

- userAgent: mozilla/5.0 (macintosh; intel mac os x 10_15_7) applewebkit/537.36 (khtml, like gecko) chrome/147.0.0.0 safari/537.36
- isTouchDevice: false

Generated at: 2026-05-13T04:38:47.861Z

### Operating System

NAME="OpenCloudOS" VERSION="9.2" ID="opencloudos" ID_LIKE="opencloudos" VERSION_ID="9.2" PLATFORM_ID="platform:oc9" PRETTY_NAME="OpenCloudOS 9.2" ANSI_COLOR="0;31" CPE_NAME="cpe:/o:opencloudos:opencloudos:9" HOME_URL="https://www.opencloudos.org/" BUG_REPORT_URL="https://bugs.opencloudos.tech/"

### n8n Version

2.19.5

### Node.js Version

24.14.1

### Database

PostgreSQL

### Execution mode

main (default)

### Hosting

self hosted