[Regression] Notion Trigger fires ONLY ONCE then fails to detect subsequent events (Self-hosted QNAP)

Questions
, ,
1

I need to report a critical regression with the Notion Trigger node on my self-hosted instance (QNAP).

Important Context (Regression): This setup was working perfectly stable in versions prior to 2.0. The issue only started appearing after I updated my n8n deployment to the latest version available on QNAP (2.2.6). Before this update (in version < 2.0), the Notion Trigger would run continuously without issues.

The Critical Issue (One-and-Done Failure): The Notion Trigger (configured to poll “Every Minute”) successfully catches the first event, but fails to detect any subsequent events.

  1. Scenario A (Restart Container): If I restart the Docker container (docker restart n8n), and then add a new page to Notion, the trigger successfully runs once. However, if I continue to add more pages to Notion afterwards, it does not trigger anymore. It simply stops detecting new data after the first run.

  2. Scenario B (Republish Workflow): If I click “Publish” (Re-activate) in the UI, the same behavior occurs:

  • I add Page A → Trigger fires (Success).

  • I add Page B → Trigger does NOT fire (Failure).

It seems like the polling mechanism works for the initial check but fails to maintain the loop for subsequent checks.

Error Message: "The connection to the server was closed unexpectedly, perhaps it is offline."

{
“nodes”: [
{
“parameters”: {
“pollTimes”: {
“item”: [
{
“mode”: “everyMinute”
}
]
},
“databaseId”: {
“__rl”: true,
“value”: “2c089e25a8a7806588b3efbe4267ed0d”,
“mode”: “id”
},
“event”: “pageAddedToDatabase”
},
“type”: “n8n-nodes-base.notionTrigger”,
“typeVersion”: 1,
“position”: [
912,
-720
],
“id”: “9059c40b-38ad-4123-99c7-4579d0fb9ec6”,
“name”: “Notion Trigger”,
“retryOnFail”: true,
“credentials”: {
“notionApi”: {
“id”: “ykTauw7XkVFGQhF8”,
“name”: “Notion account”
}
}
}
],
“connections”: {}
}

Information on your n8n setup

  • n8n version: 2.2.6 (Package version on QNAP Container Station)

  • Database (default: SQLite): PostgreSQL

  • n8n EXECUTIONS_PROCESS setting (default: own, main): Default

  • Running n8n via (Docker, npm, n8n cloud, desktop app): Docker (QNAP Container Station)

  • Operating system: QNAP QTS (Linux)

2

I looked into the Notion Trigger node and I think I found the issue. You need to update the `pollTimes` setting in your Vector Store Retriever node to “every 5 minutes” instead of just every minute. This will ensure that the trigger fires again after the initial poll, which is what’s happening right now.

```json

{

“timeout”: 30000,

“topK”: 5,

“pollTimes”: {

"item": \[

  {

    "mode": "every 5 minutes"

  }

\]

}

}

```

Try updating this setting and see if it resolves the issue. If you’re still having trouble, feel free to let me know!