Twitter node again

Hi everyone,
So I do have a self hosted version of n8n I updraded to 1.9.3
Due to recent changes on X/Twitter, many of my old nodes are not working anymore, so I reverted to this new version to update my workflows.
I recreated the App on the X developer site, just to be sure ther isn’t any issues in there.
So I wanted to use this thread to get things straight as I saw so many question from other people (me included) regarding Twitter which it seems to be an issue here.

1- What are we supposed to use as protocol now, OAuth or OAuth2
2- Is the X node working now, because I still have difficulties in getting it to work (Authorization URL issues)
3- Are we supposed to use API Keys or Customer Keys from X developer platform to use the X node ?

At the moment, none of the alternatives work for me anymore, I used to use HTTP node, but it is not working anymore simply because the Credentials are not workin on Twitter anymore, and this is so frustrating from user perspective ( I know it is not n8n, it is Twitter)

Can we have a reference guide, to clean up a bit the confusion that all of us have here about the latest changes of Twitter and how we can deal with that, I know they blocked all OAuth1 can we have a confirmaiton here ? ALso the Ofiicial n8n documentation is referencing API Key and Secretes to be used, this is actually not working for reference



Hey @Rad,

The quick version is if you are using the new node we use OAuth2 now which works with the newer API versions.

When you try to authenticate what error are you seeing?

Hey @Jon ,

So I am getting this screen first :slight_smile:

Then I get a 404 page

after the redirect is called

And I checked everything is correctly done many times according to the documentation

Hey @Rad,

That last one is interesting, Are you using the tunnel option? I suspect the tunnel crashed and if you restart n8n you will be given a new URL that will work.

I would also recommend not using the tunnel option in a production environment because this sort of thing can happen.

Hi @Jon hmm yes I am using the tunnel version, do you have any link to an example docker-compose for production setup ?

That’s my compose file

version: '3.8'


    image: postgres:11
    restart: always
      - db_storage:/var/lib/postgresql/data
      - ./
      test: ['CMD-SHELL', 'pg_isready -h localhost -U ${POSTGRES_USER} -d ${POSTGRES_DB}']
      interval: 5s
      timeout: 5s
      retries: 10

    image: n8nio/n8n
    mem_limit: 512m
    mem_reservation: 128M
    cpus: 0.5
    restart: always
      - DB_TYPE=postgresdb
      - DB_POSTGRESDB_HOST=postgres
      - N8N_BASIC_AUTH_ACTIVE=true
      - 5678:5678
      - postgres
      - n8n_storage:/home/node/.n8n
    command: "start --tunnel"
        #command: /bin/sh -c "n8n start --tunnel"
        condition: service_healthy

Ok @Jon a restart did solve the problem indeed, so what is the alternative to the tunnel version ?

Hey @Rad,

The alternative would be to set up a DNS record that uses your own domain and configure any firewalls needed for your environment. I think most of our install guides briefly mention it but as it can be very different depending on what you are using we don’t have detailed instructions.

There are also other tunnel services like Cloudflares that can be used, Assuming you are running on a VPS or similar somewhere that is already available on the internet it should just be a case of pointing an A record to your server and if you have a reverse proxy running routing it through.

1 Like

Problem solved @Jon
thanks for your help

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.