Slack Modal responses - not able to receive on an instance of N8N

Questions
1

Hi,

I’m using N8N on my home Docker instance.

I’ve been battling an issue where I’m not able to receive any window/modal response to N8N from Slack for a number of months now without any luck at all.

To be clear: I’m able to receive clicks on buttons in Slack etc. It’s just when I get N8N to create a Window in Slack a am expecting the reply - no message ever comes.

Now I’ve tried to setup a new Docker container with a new instance of N8N for a friend and test it with him and it worked just fine.

I’ve then tried to create a copy of my docker container and setup a new integration - still no luck.

I have then created a new container again for myself and there it works fine.

Any idea what the issue could be? I’d like to fix it in my current instance, please.

Please let me know what information you need from me.

Running the docker container on a Synology NAS using Portainer since ~ 2022

2

Hi @Martin_Horacek
Most likely: The Slack node in your long-running workflow is an outdated version (legacy v1/v2) that handles modal payloads differently than the fresh instance.

Open the failing workflow in your original instance. • Delete the existing Slack node entirely from the canvas (don’t just update it). • Add a brand new Slack node from the panel to ensure you get the latest version (v2+). • Re-connect the credentials and modal configuration manually. • Execute the workflow and test the modal again.

3

I’ve tried that. Did not work,

4

Can you Share any error messages or log entries from your problematic n8n instance.
and What are the N8N_HOST and WEBHOOK_URL environment variables in your problematic n8n instance?
Compare N8N_HOST and any proxy environment variables between your working and non-working n8n instances

5

@Martin_Horacek is there any logs you can show us? Is your instance setup to where public services can acsess it?

6

Hi @Martin_Horacek, welcome back to the n8n comnuty. Slack modals are stricter. In long running setups like yours, this usually comes down to public URL and proxy configuration. The WEBHOOK_URL must be explicitly set, and N8N_HOST and N8N_PROTOCOL need to match exactly how Slack reaches your n8n instance, so Slack can validate the endpoint and n8n can correctly correlate the modal submission.

7

That’s my main issue.
There are no logs at all that I could see. There’s nothing in the docker log. Are there any other logs I could find?

8

My public IP/URL and proxy configs are the same across my old instance, the new copy with new cloudflare setup etc. and new instance with also new cloudflare setup.
It really seems to be an issue with some internal N8N config.

9

There are no errors at all. Nothing reaches N8N.

10

@Martin_Horacek

Have you already exported the workflow as JSON removed the old workflow entirely restarted the container and imported it back? If this doesn’t work then it is very likely related to the N8N_ENCRYPTION_KEY or a corrupted database index.

11

I’ve done more than that as you can read above.

I’ve recreated the workflow in a copied instance manually and still the same issue.

This seems to be some background setting issue.