Why isn't the Connect triggers webhook reaching my backend application despite showing up in my queries?

This topic was automatically generated from Slack. You can find the original thread here.

Hi, I’m trying to use Connect triggers to call a webhook endpoint in my backend application. However, more often than not, the webhook never reaches my backend. Interestingly, I can see the events that should have triggered the webhook when I query the /events endpoint of my deployed trigger, and when I query /webhooks, my webhook is listed. But nothing happens. I may be misconfiguring something—could someone take a quick look?

Output of /v1/connect/<project_id>/deployed-triggers/<deployed_trigger_id>:

{
   "data" : {
      "active" : true,
      "component_id" : "sc_x0i3yNe",
      "configurable_props" : [
         {
            "app" : "google_calendar",
            "name" : "googleCalendar",
            "type" : "app"
         },
         {
            "name" : "db",
            "type" : "$.service.db"
         },
         {
            "default" : [
               "primary"
            ],
            "description" : "Select one or more calendars to watch (defaults to the primary calendar)",
            "label" : "Calendars",
            "name" : "calendarIds",
            "optional" : true,
            "remoteOptions" : true,
            "type" : "string[]"
         },
         {
            "default" : false,
            "description" : "Emit new events only, and not updates to existing events (defaults to `false`)",
            "label" : "Emit only for new events",
            "name" : "newOnly",
            "optional" : true,
            "type" : "boolean"
         },
         {
            "name" : "http",
            "type" : "$.interface.http"
         },
         {
            "description" : "The Google Calendar API requires occasional renewal of push notification subscriptions. ****This runs in the background, so you should not need to modify this schedule****.",
            "label" : "Push notification renewal schedule",
            "name" : "timer",
            "static" : {
               "intervalSeconds" : 82080
            },
            "type" : "$.interface.timer"
         }
      ],
      "configured_props" : {
         "calendarIds" : [
            "primary"
         ],
         "db" : {
            "type" : "$.service.db"
         },
         "googleCalendar" : {
            "authProvisionId" : "apn_xxx" // redacted
         },
         "http" : {
            "endpoint_url" : "https://xxx.m.pipedream.net" // redacted
         },
         "newOnly" : false,
         "timer" : {
            "cron" : null,
            "interval_seconds" : 82080
         }
      },
      "created_at" : 1750835973,
      "id" : "xxx", // redacted
      "name" : "xxx", // redacted
      "name_slug" : "xxx", // redacted
      "owner_id" : "exu_MLiKQ4",
      "updated_at" : 1750835973
   }
}

can you help?

Absolutely, I can help! Based on your description, you’re using Pipedream Connect triggers to call a webhook endpoint in your backend, but the webhook is not being delivered, even though you see the events in /events and your webhook is listed in /webhooks. Let’s walk through common causes and troubleshooting steps for this issue: — ### 1. Check Webhook Delivery Status - Events in /events but not hitting your backend means the trigger is firing, but the webhook delivery is failing or not being attempted. - Pipedream Connect should attempt to POST to your endpoint_url for each event. If it fails, you should see delivery errors in the Connect dashboard or via the API. Action: - Check the Connect dashboard for your deployed trigger for any error logs or failed delivery attempts. - If available, use the Connect API to inspect the delivery status of recent events. — ### 2. Verify Webhook Configuration From your config: on "http": { "endpoint_url": "https://xxx.m.pipedream.net" } - Make sure the endpoint_url is correct and publicly accessible. - If you’re using a Pipedream workflow as your webhook endpoint, ensure that workflow is deployed and has an HTTP trigger. — ### 3. Check for Delivery Failures - If your endpoint returns a non-2xx status code, Pipedream Connect may retry or mark the delivery as failed. - If your endpoint is slow to respond or times out, delivery may also fail. Action: - Check your backend logs for any incoming requests or errors. - Temporarily replace the endpoint_url with a service like Webhook.site to confirm if deliveries are attempted at all. — ### 4. Confirm Trigger Activation Your trigger shows "active": true, so it should be running. But if you recently updated the webhook or trigger, try deactivating and reactivating the trigger to refresh its state. — ### 5. Check for Rate Limiting or Filtering - Some triggers have deduplication or event filtering logic. If you set "newOnly": false, you should get all events, but double-check if your backend is filtering out events or if Connect is suppressing duplicates. — ### 6. Test with a Simple Endpoint - Use a simple HTTP endpoint (like Webhook.site or a basic Express server) to confirm that Pipedream Connect is making the POST requests. — ### 7. Contact Support If you’ve checked all the above and still have issues, there may be a bug or an internal delivery problem. Please visit Support - Pipedream for more support options. — ## Summary Checklist - Confirm endpoint_url is correct and accessible. - Check Connect dashboard/API for delivery errors. - Inspect backend logs for incoming requests. - Test with a known-good endpoint (Webhook.site). - Deactivate/reactivate the trigger if recently changed. - ] Contact [Pipedream Support if the issue persists. Let me know if you need help with any specific step or want to share more details (like error logs, delivery status, or endpoint code)!

, quick question, how is it configured to trigger your backend? The http.endpoint_url is a Pipedream webhook URL, right?

Hi Jup it’s a pipedream webhook url the xxx is actually some random string but redacted it for security reasons.

I think it should be like this right? Eventhough I don’t configure this url myself. I use this endpoint to set it up:

POST: /triggers/deploy
{
    "id": "google_calendar-new-or-updated-event-instant",
    "configured_props": {
        "googleCalendar": {
            "authProvisionId": "redacted"
        },
        "calendarIds": ["primary"],
        "newOnly" : false
    },
    "external_user_id": "redacted",
    "webhook_url": "https://redacted.ngrok-free.app/webhook",
}

any ideas? This challenge is blocking us from rolling out a new feature to our customers

Anybody else from the pipedream team? Or should I make a support ticket for this issue? Happy to solve it myself, but wouldn’t know where to look for potential errors during webhook delivery.

CC: , @U03CXTHK4RF

Andrew / Leo lmk if you guys are able to help debug here :pray::skin-tone-2:

It seems to be a bug in the trigger, I see it’s getting triggered from Google but not emitting events

Wait nvm

I’ll follow-up in the support ticket