Connection lost: You have a connection issue or the server is down. n8n should reconnect automatically once the issue is resolved

[fritz_gonzales]

this worked!! love you man, added -e N8N_PUSH_BACKEND=sse in dokcer run

contributing my docker run here for future reference

make volume n8n_data
and make network n8n_network

docker run -d --name n8n --network n8n_network --dns=8.8.8.8 --dns=1.1.1.1 --dns-search=domain.com --e N8N_HOST="subdomain.domain.com" --e WEBHOOK_URL="https://subdomain.domain.com/" --e VUE_APP_URL_BASE_API="https://subdomain.domain.com/" -e NODE_FUNCTION_ALLOW_EXTERNAL="*" -e N8N_PUSH_BACKEND=sse -v n8n_data:/home/node/.n8n -p 5678:5678 n8nio/n8n

just replace subdomain and domain values

[Mohamed_Saleh1]

Folks using NGINX Proxy Manager, all you have to do is to enable Websockets Support.

[Claudi]

After apply the N8N_PUSH_BACKEND=see, under the logs I see the following message:

Collaboration features are disabled because push is configured unidirectional. Use N8N_PUSH_BACKEND=websocket environment variable to enable them.
Version: 1.86.1

What does it mean? I have activated the Websockets Support under my proxy manager

[Josia_Almeida]

I enabled the Webhook on my Nginx and it passed through correctly. I did the same process for my API (EvolutionAPI).
That solution worked for me.

[NLSolutions]

I actually got rid of this issue in the past, but now I downloaded the next (1.88) version and the issue is showing itself again. However, I cloned the container that worked (using docker on synology ds923), but I keep getting the Connection lost issue. Anyone else facing that switching from latest to next version?

[NLSolutions]

Here’s the YML btw:

version: ‘3.8’

services:
n8n:
image: docker.n8n.io/n8nio/n8n:next
container_name: n8n-next
restart: always
ports:
- “5678:5678”
environment:
- TZ=Europe/Amsterdam
- N8N_BASIC_AUTH_ACTIVE=true
- N8N_BASIC_AUTH_USER=MYUSER
- N8N_BASIC_AUTH_PASSWORD=MYPASSWORD
- N8N_PORT=5678
- N8N_HOST=n8n.northernlights.solutions
- N8N_PROTOCOL=https
- N8N_EDITOR_BASE_URL=https://n8n.northernlights.solutions
- WEBHOOK_URL=https://n8n.northernlights.solutions/
- N8N_SKIP_WEBHOOK_DEREGISTRATION_SHUTDOWN=true
- N8N_EXECUTIONS_PROCESS=main
- N8N_RUNNERS_ENABLED=true
- N8N_PUSH_BACKEND=sse
- N8N_SECURE_COOKIE=false
volumes:
- /volume1/NorthernLights Solutions/n8n/data:/home/node/.n8n
- /volume1/NorthernLights Solutions/n8n/files:/files

(this exact config works perfectly on ‘latest’)

[johnnywaffles]

I’m getting that error again as well. It just started showing up in the past couple of days. I’m using “sse” as the N8N_PUSH_BACKEND value.

[Alireza_Nasiri]

Changing nginx config to this fixed my problem, hope it works for you too:

server {
    server_name yourdomain;

    location / {
        proxy_pass http://localhost:5678;
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection $connection_upgrade;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;

        proxy_buffering off;
        chunked_transfer_encoding off;
        proxy_cache off;
    }
}

map $http_upgrade $connection_upgrade {
    default upgrade;
    '' close;
}

Also if you use cloudflare, disabling caching for your subdomain is recommended.

[Douglas_Lopes]

Yes I had the same issue when migrating to 1.88. Opened a issue in github: Connection lost when behind a CDN · Issue #14572 · n8n-io/n8n · GitHub

[NLSolutions]

For me the issue was solved switching the SSE to websocket in the N8N_PUSH_BACKEND environment. (and enable proper websocketing settings on my Synology DS923 NAS)

[tgkamal]

in nginx .conf

/location
add this code

        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection 'upgrade';
        proxy_set_header Host $host;
        proxy_cache_bypass $http_upgrade;

in env docker must add N8N_PUSH_BACKEND=websocket

this is my solution for Version: 1.88.0

[Bernard_Verhaeghe]

I got it to work by putting the script hereunder in the nginx manager and changing the manager and N8N_PUSH_BACKEND both to websocket. But after that I got enough of this recurrung problem and with the prompt help of claude.ai I got a YML script for traefik and in no time I eliminated the use of nginx once and for all.

Anyway, this was the solution with nginx:

location /sockjs-node {
proxy_pass https://n8n:5678;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection “upgrade”;
proxy_set_header Host $host;
proxy_cache_bypass $http_upgrade;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}

location /websocket {
proxy_pass https://n8n:5678;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection “upgrade”;
proxy_set_header Host $host;
proxy_cache_bypass $http_upgrade;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}

Kind regards

[Jim_Shingler]

This worked for me: Connection lost after upgrade to 1.88.0 · Issue #14653 · n8n-io/n8n · GitHub

[Sergio_Vega]

Greetings, everyone:

I had the exact same issue. In my case, since I’m using Nginx Proxy Manager, the solution was to enable the “Websockets Support” option in the proxy host settings.

[Qu_c_Thai]

I have successfully fixed it!
“Connection lost” error happens from n8n version 1.79.3 and above.
Solution: downgrade to n8n 1.78.1 + fix WebSocket proxy settings in Nginx.
Then restart Nginx — the issue will be resolved.

[BramKn]

Solution as given is not perse the right one, as reverting to a previous version should not be needed.
As mentioned by multiple people replying here, the issue is that WSS needs to be enabled in the reverse proxy.

[marcelo.magalhaes]

@TofDevNet

N8N_P

Many thanks for your tip

I’m running 1.90.2 and I had to add this line to the nginx conf file:

proxy_set_header Origin https://$host;

[Ricardo_Aveiro]

This configuration works for me on 1.90 with N8N_PUSH_BACKEND=websocket and N8N_PROXY_HOPS=1 environment variable

[Alexandre_Correa]

Hi, guys.
I upgraded from version 1.78.1 to 1.90.1 and had several problems. To help you out, I’ll share what worked for me:

First, upgrade through the versions.
I went to 1.80.1
in NGINX:
proxy_http_version 1.1; # Required for WebSocket
proxy_set_header Upgrade $http_upgrade; # Required for WebSocket
proxy_set_header Connection “upgrade”; # Required for WebSocket

N8N:
N8N_ENFORCE_SETTINGS_FILE_PERMISSIONS=true
N8N_RUNNERS_ENABLED=true
VUE_APP_URL_BASE_API=“https://DOMINIO.COM”

Command line used:
docker run -d -it --rm --name n8n -p 5678:5678 -e N8N_RUNNERS_ENABLED=true -e N8N_ENFORCE_SETTINGS_FILE_PERM
ISSIONS=true -e N8N_PUSH_BACKEND=websocket -e N8N_PROXY_HOPS=1 -e VUE_APP_URL_BASE_API=“https://DOMINIO.COM” -e WEBHOOK_URL=https://DOMINIO.COM -v n8n_data:/home/node/.n8n n8nio/n8n:latest

Then upgrade to:
1.90.1

Here is my command line as a final example after running
docker run -d -it --rm --name n8n -p 5678:5678 -e WEBHOOK_URL=https://DOMINIO.COM -v n8n_data:/home/node/.
n8n docker.n8n.io/n8nio/n8n

Thanks to everyone here because with your help I was able to do everything.