[Solved] Notion Issues with [email protected]

Hi Guys,

When trying to update the .env file and restart docker-compose, I stuffed something up and am running into problems with n8n:

  • Existing flows have now lost a lot of modules (a few are left and are disconnected)
  • I get the following error everywhere:

I have all workflows backed up on git, so can restore them, but seems that there is some error with n8n.

Would be very grateful for any assistance

Hi @pb84, could you describe which changes you have made? How do your .env and docker-compose.yml files looks like and what does your server log (docker-compose logs) say?

You can additionally also set the environment variable N8N_LOG_LEVEL to debug to get additional information displayed.

Thank you @MutedJam and @jan for getting back to me so quickly.

What I did:

  • I was trying to make n8n run faster by adding export EXECUTIONS_PROCESS=main to the .env file
  • When doing so, I think I incorrectly restarted docker-compose to initiate the changes by following some advice on the community without carefully understanding it:

Since this happened I have:

  • Removed the addition from the .env file and have restarted docker with what I believe is the correct method (docker-compose down, docker-compose up -d)
  • Updated n8n to the latest version (the one released today, was previously on the version before).

I believe I have changed something to port 5678 (visible from the logs):

  • Note: I have just blurred out some stuff that isn’t important, including domain and account details on some of the workflow names. If you need anything from above let me know.

@jan, will get the other logs that you’ve mentioned if above doesn’t fix, but i think its to do with the 5678 issue

Thank you again so much, really appreciated

Is there a chance your manually started docker container is still running and occupying port 5678, preventing your docker compose-configured container from launching as expected?

What is the output of docker ps on your server?

That’s the output from docker ps

What used to look like this:

Now looks like this:

So I don’t think that’s a problem with the specific ports. But let’s take it step by step.

Once we know which environment variables you are setting and how your current docker compose file looks like we’re hopefully able to at get an idea if how your environment looks like.

The server logs should give us an indicator of whether there’s something wrong with the server-side of things. In a next step you could check if your browser can reach whatever n8n is throwing at it or if there’s any failing requests, perhaps blocked in error by traefik (this page describes how to check your browser console).

1 Like

Could you also share the network tab of your browser console? Just to see if there’s any problems with loading any of n8n’s components?

Sure

I think I have an idea of what is going on. Seems to be related to the Notion node v1. Will try to find a fix asap.

1 Like

Yep, interestingly, flow still work in the background (e.g. have something with webhooks thats running smoothly, just the GUI isn’t looking nice)

Also, yep, all flows which don’t have notion are looking fine

1 Like

Side question - Is it possible to downgrade n8n, e.g. revert back to version 0.174.1?

Yes, the fastest solution would be to downgrade to [email protected]

The docker image of version [email protected] is currently building which includes the bugfix.

But unless there is anything in that version that you need, you can simply keep on running on [email protected]. Here the changelog.

OK [email protected] is now available.

3 Likes