I implemented the parallel pattern as described in the n8n documentation (using callbacks). It works fine, except when I call the main flow using a tool that is called by an AI agent.
What happens is that the respond to webhook output is returned immediately back to the AI agent, so there is effectively no wait for the main flow to finish. And I need the agent to wait for the workflow to finish, before continuing.
What am I missing? Is this a bug, or is there a possible workaround?
The workflows:
{
“nodes”: [
{
“parameters”: {
“options”: {}
},
“type”: “@n8n/n8n-nodes-langchain.chatTrigger”,
“typeVersion”: 1.1,
“position”: [
0,
0
],
“id”: “cd5d4c62-b47b-4647-a313-ff5cf10a41f5”,
“name”: “When chat message received”,
“webhookId”: “5b8d6b75-6ab4-42a9-a9eb-005f2bb7161f”
},
{
“parameters”: {
“options”: {}
},
“type”: “@n8n/n8n-nodes-langchain.agent”,
“typeVersion”: 1.8,
“position”: [
220,
0
],
“id”: “6b3b8cdd-8441-4681-9787-0cffa597bc50”,
“name”: “AI Agent”
},
{
“parameters”: {
“model”: “gpt-4o-mini”,
“options”: {}
},
“type”: “@n8n/n8n-nodes-langchain.lmChatAzureOpenAi”,
“typeVersion”: 1,
“position”: [
260,
220
],
“id”: “942c573e-1f95-4108-81b5-69ee54318971”,
“name”: “Azure OpenAI Chat Model”,
“credentials”: {
“azureOpenAiApi”: {
“id”: “iKY0wB5wcFToSq7t”,
“name”: “Azure gpt-40-mini”
}
}
},
{
“parameters”: {
“name”: “pp_tool”,
“workflowId”: {
“__rl”: true,
“value”: “iue5rBXxnFXZd8MK”,
“mode”: “list”,
“cachedResultName”: “Parallel Exec with wait MAIN flow - template”
},
“workflowInputs”: {
“mappingMode”: “defineBelow”,
“value”: {},
“matchingColumns”: ,
“schema”: ,
“attemptToConvertTypes”: false,
“convertFieldsToString”: false
}
},
“type”: “@n8n/n8n-nodes-langchain.toolWorkflow”,
“typeVersion”: 2.1,
“position”: [
380,
220
],
“id”: “334d73cc-1568-47a0-8a45-89e2d8560be1”,
“name”: “Call n8n Workflow Tool”
}
],
“connections”: {
“When chat message received”: {
“main”: [
[
{
“node”: “AI Agent”,
“type”: “main”,
“index”: 0
}
]
]
},
“Azure OpenAI Chat Model”: {
“ai_languageModel”: [
[
{
“node”: “AI Agent”,
“type”: “ai_languageModel”,
“index”: 0
}
]
]
},
“Call n8n Workflow Tool”: {
“ai_tool”: [
[
{
“node”: “AI Agent”,
“type”: “ai_tool”,
“index”: 0
}
]
]
}
},
“pinData”: {},
“meta”: {
“instanceId”: “58aa91dfb81527f3ab9d46e751edfcf1e8801ddcdd145e7779c36f37dcdfa475”
}
}
{
“nodes”: [
{
“parameters”: {
“jsCode”: “// an example item for the loop/split out\nconst myData = [\n {id: 1, content: ‘String one’},\n {id: 2, content: ‘String two’}\n];\nreturn myData;”
},
“type”: “n8n-nodes-base.code”,
“typeVersion”: 2,
“position”: [
280,
0
],
“id”: “2b3aa72b-771a-45a6-8ab3-9a4182d95e2f”,
“name”: “Generate item with x objects”
},
{
“parameters”: {
“options”: {}
},
“type”: “n8n-nodes-base.splitInBatches”,
“typeVersion”: 3,
“position”: [
500,
0
],
“id”: “dc3135df-414b-4d94-a92c-d10f29df2b49”,
“name”: “Loop Over Items”
},
{
“parameters”: {
“resume”: “webhook”,
“httpMethod”: “POST”,
“responseMode”: “responseNode”,
“options”: {}
},
“type”: “n8n-nodes-base.wait”,
“typeVersion”: 1.1,
“position”: [
1340,
120
],
“id”: “04bbb998-967d-4b7d-b0ec-315acb04efa0”,
“name”: “Listen for subflow callback”,
“webhookId”: “1e99daa4-e050-4a46-aba5-8b68bcfc6e49”
},
{
“parameters”: {
“method”: “POST”,
“url”: “https://n8n-host/webhook/parallel-webhook-call”,
“sendHeaders”: true,
“headerParameters”: {
“parameters”: [
{
“name”: “=callbackurl”,
“value”: “={{ $execution.resumeUrl }}”
}
]
},
“sendBody”: true,
“bodyParameters”: {
“parameters”: [
{
“name”: “id”,
“value”: “={{ $json.id }}”
}
]
},
“options”: {}
},
“type”: “n8n-nodes-base.httpRequest”,
“typeVersion”: 4.2,
“position”: [
720,
160
],
“id”: “eccbaac8-94e0-48bb-ae6f-a435f6a3ac8b”,
“name”: “HTTP call subflow”
},
{
“parameters”: {
“jsCode”: “let result = $(‘Listen for subflow callback’).first().json.body;\n\nlet json = $(‘test endResult cnt’).first().json;\nif (!json.endResult) json.endResult = ; //init\n\njson.endResult.push(result);\n\nreturn [json];”
},
“type”: “n8n-nodes-base.code”,
“typeVersion”: 2,
“position”: [
1520,
120
],
“id”: “6928941d-f850-47ed-bf13-6c4b140b86ce”,
“name”: “Add execution result”
},
{
“parameters”: {
“conditions”: {
“options”: {
“caseSensitive”: true,
“leftValue”: “”,
“typeValidation”: “strict”,
“version”: 2
},
“conditions”: [
{
“id”: “99603d4a-4e96-4df9-b1bd-4f1cda6d8444”,
“leftValue”: “={{ $json.endResult.length }}”,
“rightValue”: “={{ $(‘Generate item with x objects’).all().length }}”,
“operator”: {
“type”: “number”,
“operation”: “equals”
}
}
],
“combinator”: “and”
},
“options”: {}
},
“type”: “n8n-nodes-base.if”,
“typeVersion”: 2.2,
“position”: [
1120,
40
],
“id”: “b2419ac4-fdea-4b7b-a7a2-e15903ced7da”,
“name”: “test endResult cnt”
},
{
“parameters”: {
“options”: {}
},
“type”: “n8n-nodes-base.respondToWebhook”,
“typeVersion”: 1.1,
“position”: [
1700,
120
],
“id”: “f6d4ce81-a994-45b1-abac-e45518e98b7a”,
“name”: “Ack subflow”
},
{
“parameters”: {},
“type”: “n8n-nodes-base.wait”,
“typeVersion”: 1.1,
“position”: [
1700,
-120
],
“id”: “52dcc50f-08be-4648-9c63-2f416b6afca5”,
“name”: “Yes, we made it”,
“webhookId”: “af3a03a0-5276-4358-b21d-4d7975171265”
},
{
“parameters”: {
“content”: “## Call parallel subflows \nUses http call node",
“height”: 640,
“width”: 940
},
“type”: “n8n-nodes-base.stickyNote”,
“typeVersion”: 1,
“position”: [
-40,
-200
],
“id”: “b50ada1d-4abf-4b57-b364-cf12e74ebe04”,
“name”: “Sticky Note”
},
{
“parameters”: {
“content”: "## Wait for subflows to report back in\n jib”,
“height”: 640,
“width”: 940
},
“type”: “n8n-nodes-base.stickyNote”,
“typeVersion”: 1,
“position”: [
1000,
-200
],
“id”: “a197a4ec-04b7-4a78-a8bf-937b43871e7d”,
“name”: “Sticky Note1”
},
{
“parameters”: {
“options”: {}
},
“type”: “n8n-nodes-base.respondToWebhook”,
“typeVersion”: 1.1,
“position”: [
1920,
-120
],
“id”: “8c14a086-7389-4bcc-8166-b1c080707231”,
“name”: “Respond to Webhook”
},
{
“parameters”: {
“inputSource”: “passthrough”
},
“type”: “n8n-nodes-base.executeWorkflowTrigger”,
“typeVersion”: 1.1,
“position”: [
0,
0
],
“id”: “efc1986a-12c6-46bb-a013-22e3c528867a”,
“name”: “When Executed by Another Workflow”
}
],
“connections”: {
“Generate item with x objects”: {
“main”: [
[
{
“node”: “Loop Over Items”,
“type”: “main”,
“index”: 0
}
]
]
},
“Loop Over Items”: {
“main”: [
[
{
“node”: “test endResult cnt”,
“type”: “main”,
“index”: 0
}
],
[
{
“node”: “HTTP call subflow”,
“type”: “main”,
“index”: 0
}
]
]
},
“Listen for subflow callback”: {
“main”: [
[
{
“node”: “Add execution result”,
“type”: “main”,
“index”: 0
}
]
]
},
“HTTP call subflow”: {
“main”: [
[
{
“node”: “Loop Over Items”,
“type”: “main”,
“index”: 0
}
]
]
},
“Add execution result”: {
“main”: [
[
{
“node”: “Ack subflow”,
“type”: “main”,
“index”: 0
}
]
]
},
“test endResult cnt”: {
“main”: [
[
{
“node”: “Yes, we made it”,
“type”: “main”,
“index”: 0
}
],
[
{
“node”: “Listen for subflow callback”,
“type”: “main”,
“index”: 0
}
]
]
},
“Ack subflow”: {
“main”: [
[
{
“node”: “test endResult cnt”,
“type”: “main”,
“index”: 0
}
]
]
},
“Yes, we made it”: {
“main”: [
[
{
“node”: “Respond to Webhook”,
“type”: “main”,
“index”: 0
}
]
]
},
“When Executed by Another Workflow”: {
“main”: [
[
{
“node”: “Generate item with x objects”,
“type”: “main”,
“index”: 0
}
]
]
}
},
“pinData”: {},
“meta”: {
“instanceId”: “58aa91dfb81527f3ab9d46e751edfcf1e8801ddcdd145e7779c36f37dcdfa475”
}
}
{
“nodes”: [
{
“parameters”: {
“jsCode”: “// Loop over input items and add a new field called ‘myNewField’ to the JSON of each one\nfor (const item of $input.all()) {\n item.json.addition = ‘Data added by subflow’;\n}\n\nreturn $input.all();”
},
“type”: “n8n-nodes-base.code”,
“typeVersion”: 2,
“position”: [
440,
0
],
“id”: “012a8cb9-a387-48ba-bc37-e0bec572f884”,
“name”: “Generate subflow data (do stuff)”
},
{
“parameters”: {
“httpMethod”: “POST”,
“path”: “parallel-webhook-call”,
“responseMode”: “responseNode”,
“options”: {}
},
“type”: “n8n-nodes-base.webhook”,
“typeVersion”: 2,
“position”: [
0,
0
],
“id”: “1c94229e-35de-4fe4-ae33-6cd907667d6d”,
“name”: “Webhook”,
“webhookId”: “b115cd5b-8624-47fb-a710-a3fd08aa8326”
},
{
“parameters”: {
“method”: “POST”,
“url”: “={{ $(‘Webhook’).item.json.headers.callbackurl }}”,
“sendBody”: true,
“bodyParameters”: {
“parameters”: [
{
“name”: “returnStatus”,
“value”: “=1”
},
{
“name”: “id”,
“value”: “={{ $(‘Webhook’).item.json.body.id }}”
}
]
},
“options”: {}
},
“type”: “n8n-nodes-base.httpRequest”,
“typeVersion”: 4.2,
“position”: [
900,
0
],
“id”: “d9305549-d610-4307-a66d-1afda9155963”,
“name”: “Send result back to main flow”,
“retryOnFail”: true
},
{
“parameters”: {
“options”: {}
},
“type”: “n8n-nodes-base.respondToWebhook”,
“typeVersion”: 1.1,
“position”: [
220,
0
],
“id”: “f8939365-0cfd-4159-8e3e-6f34f98dddac”,
“name”: “Respond to Webhook”
},
{
“parameters”: {
“amount”: “={{ Math.ceil(Math.random() * (5 - 2) + 2) }}”
},
“type”: “n8n-nodes-base.wait”,
“typeVersion”: 1.1,
“position”: [
660,
0
],
“id”: “666119fe-9cf7-41fa-9e71-b3ea5be1fb91”,
“name”: “Wait random 2-5 secs”,
“webhookId”: “d4515c9b-5b72-47d8-af36-550dd58e103e”
}
],
“connections”: {
“Generate subflow data (do stuff)”: {
“main”: [
[
{
“node”: “Wait random 2-5 secs”,
“type”: “main”,
“index”: 0
}
]
]
},
“Webhook”: {
“main”: [
[
{
“node”: “Respond to Webhook”,
“type”: “main”,
“index”: 0
}
]
]
},
“Send result back to main flow”: {
“main”: [
]
},
“Respond to Webhook”: {
“main”: [
[
{
“node”: “Generate subflow data (do stuff)”,
“type”: “main”,
“index”: 0
}
]
]
},
“Wait random 2-5 secs”: {
“main”: [
[
{
“node”: “Send result back to main flow”,
“type”: “main”,
“index”: 0
}
]
]
}
},
“pinData”: {},
“meta”: {
“instanceId”: “58aa91dfb81527f3ab9d46e751edfcf1e8801ddcdd145e7779c36f37dcdfa475”
}
}
Information on your n8n setup
- n8n version: Version 1.88.0
- Database: sqlite
- n8n EXECUTIONS_PROCESS default:
- Running n8n via Docker compose:
- Operating system: Debian Linux 12
