Issue install N8N in docker and server no internet connection

Questions
1

I have one problem about install n8n in server no internet connection and this my env:
environment:
- DB_TYPE=postgresdb
- DB_POSTGRESDB_HOST=postgres
- DB_POSTGRESDB_PORT=5432
- DB_POSTGRESDB_USER=${POSTGRES_USER}
- DB_POSTGRESDB_PASSWORD=${POSTGRES_PASSWORD}

# Security & Encryption
- N8N_ENCRYPTION_KEY=${N8N_ENCRYPTION_KEY}
- N8N_USER_MANAGEMENT_JWT_SECRET=${N8N_USER_MANAGEMENT_JWT_SECRET}

# External Services (Optional AI connection)
- OLLAMA_HOST=ollama:11434  # Remove if not used

# Disable Telemetry & Tracking
- N8N_DIAGNOSTICS_ENABLED=false
- N8N_TELEMETRY_ENABLED=false
- N8N_PERSONALIZATION_ENABLED=false
- N8N_VERSION_NOTIFICATIONS_ENABLED=false
- N8N_TEMPLATES_ENABLED=false

# Disable Public API Exposure
- N8N_PUBLIC_API_DISABLED=true
- N8N_PUBLIC_API_SWAGGERUI_DISABLED=true

# Disable External Tracking Hooks
- EXTERNAL_FRONTEND_HOOKS_URLS=
- N8N_DIAGNOSTICS_CONFIG_FRONTEND=
- N8N_DIAGNOSTICS_CONFIG_BACKEND=

# HTTP Config (No HTTPS Enforcement)
- N8N_PROTOCOL=http
- N8N_SECURE_COOKIE=false
- N8N_DISABLE_HTTPS_WARNING=true

# Optional Permissions Enforcement
- N8N_ENFORCE_SETTINGS_FILE_PERMISSIONS=true

And you can see the ollama model not response,I think it try to access to internet

image
image1203×833 61.9 KB

2

Hi,

Are you using the AI starter kit?

OLLAMA_HOST=ollama:11434

this would mean he would try to connect to a local container which runs OLLAMA

Reg,
J.

3

yes I follow this reo GitHub - n8n-io/self-hosted-ai-starter-kit: The Self-hosted AI Starter Kit is an open-source template that quickly sets up a local AI environment. Curated by n8n, it provides essential tools for creating secure, self-hosted AI workflows.

4

Hi,

Yes, so everything should be self-contained and work locally. I don’t see the error to be honest. you type “gy” and the AI agents is confused (so am I :). what are you trying to do.

reg,
J.

5

You see on ollama chat model it not check text when it already response

6

n8n-error
n8n-error1203×833 82.1 KB

It should reponse on this

7

Hi,

not sure to be honest.

And if you go into the credentials/account for Ollama, it says everything is ok?
All the containers are up and running ? no error state or whatever?

Have you tried to recreate the WF / start a new session?

Also: I dont know the default config but this seems weird if you are not using it

- EXTERNAL_FRONTEND_HOOKS_URLS=
- N8N_DIAGNOSTICS_CONFIG_FRONTEND=
- N8N_DIAGNOSTICS_CONFIG_BACKEND=

run the standard configuration without any modifications and check whether it works (you seem to have made modifications as compared to the repository)

regards
J.

8

Please post the logs/console output. There should be an error message that’ll help