Notes - AI Bot Mail-Chat Integration (n8n, Open WebUI)
Overview
Requirements
-
Docker / Investigations - Docker
-
open-webui/open-webui: User-friendly AI Interface (Supports Ollama, OpenAI API, ...) - https://github.com/open-webui/open-webui
Tips
-
Concurrency control | n8n Docs - https://docs.n8n.io/hosting/scaling/concurrency-control/
Logs
-
Setup Windows 10, WSL2, Docker within WSL2.
-
in WSL2
-
docker volume create n8n_data docker run -it --rm --name n8n -p 5678:5678 -v n8n_data:/home/node/.n8n docker.n8n.io/n8nio/n8n docker run -it --rm --name n8n -e N8N_SECURE_COOKIE=false -e N8N_CONCURRENCY_PRODUCTION_LIMIT=1 -p 5678:5678 -v n8n_data:/home/node/.n8n docker.n8n.io/n8nio/n8n or $ docker run -d --name n8n --restart=unless-stopped -e N8N_LOG_LEVEL=debug -e N8N_RUNNERS_ENABLED=true -e N8N_DEFAULT_BINARY_DATA_MODE=filesystem -e EXECUTIONS_DATA_PRUNE=true -e EXECUTIONS_DATA_PRUNE_MAX_COUNT=500 -e DB_SQLITE_VACUUM_ON_STARTUP=true -e NODE_OPTIONS='--max-old-space-size=2048' -e N8N_SECURE_COOKIE=false -e N8N_CONCURRENCY_PRODUCTION_LIMIT=1 -p 5678:5678 -v n8n_data:/home/node/.n8n docker.n8n.io/n8nio/n8n (RUNNER and filesystem may be unstable?) $ docker run -d --name n8n --restart=unless-stopped -e N8N_LOG_LEVEL=info -e EXECUTIONS_DATA_PRUNE=true -e EXECUTIONS_DATA_PRUNE_MAX_COUNT=500 -e DB_SQLITE_VACUUM_ON_STARTUP=true -e NODE_OPTIONS='--max-old-space-size=2560' -e N8N_SECURE_COOKIE=false -e N8N_CONCURRENCY_PRODUCTION_LIMIT=1 -p 5678:5678 -v n8n_data:/home/node/.n8n docker.n8n.io/n8nio/n8n $ docker run -d --name n8n --restart=unless-stopped -e N8N_LOG_LEVEL=debug -e EXECUTIONS_DATA_PRUNE=true -e EXECUTIONS_DATA_PRUNE_MAX_COUNT=500 -e DB_SQLITE_VACUUM_ON_STARTUP=true -e NODE_OPTIONS='--max-old-space-size=2560' -e N8N_SECURE_COOKIE=false -e N8N_CONCURRENCY_PRODUCTION_LIMIT=1 -p 5678:5678 -v n8n_data:/home/node/.n8n docker.n8n.io/n8nio/n8n $ docker run -d --name n8n --restart=unless-stopped -e N8N_LOG_LEVEL=debug -e N8N_RUNNERS_ENABLED=true -e EXECUTIONS_DATA_PRUNE=true -e EXECUTIONS_DATA_PRUNE_MAX_COUNT=500 -e DB_SQLITE_VACUUM_ON_STARTUP=true -e NODE_OPTIONS='--max-old-space-size=2560' -e N8N_SECURE_COOKIE=false -e N8N_CONCURRENCY_PRODUCTION_LIMIT=1 -p 5678:5678 -v n8n_data:/home/node/.n8n docker.n8n.io/n8nio/n8n
-
-
Using a testing gmail address.
-
Folllowing the instruction: https://docs.n8n.io/integrations/builtin/trigger-nodes/n8n-nodes-base.gmailtrigger/?utm_source=n8n_app&utm_medium=node_settings_modal-credential_link&utm_campaign=n8n-nodes-base.gmailTrigger & Google credentials | n8n Docs - https://docs.n8n.io/integrations/builtin/credentials/google/
-
Have a Google Cloud Project “N8N TEST” at https://console.cloud.google.com/
-
Prepare for enabling OAuth for the project
-
Add OAuth Client
-
Add the testing gmail address to the “Tester” in OAuth Agreement configuration.
-
-
-
-
Build n8n workflow
-
-
Configure Ollama
-
OllamaでローカルLLM - https://okumuralab.org/~okumura/misc/250418.html
-
-
e.g.
[Service] Environment="OLLAMA_ORIGINS=*" Environment="OLLAMA_KEEP_ALIVE=600" Environment="OLLAMA_RUN_TIMEOUT=600" Environment="OLLAMA_REQUEST_TIMEOUT=600" Environment="OLLAMA_HOST=0.0.0.0:11434" Environment="OLLAMA_FLASH_ATTENTION=1"
-
-
----WIP----
-
Run Install Open WebUI
-
docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
-
Task Parking Lot
-
Setup TLS/HTTPS (recommended), for tailscale network and access across instances.
-
with proper Nginx routing/hardening.
-
Enabling HTTPS · Tailscale Docs - https://tailscale.com/kb/1153/enabling-https
-
-
For testing purpose, setup a cloned instance of my Open WebUI instance: 📦 Exporting and Importing Database | Open WebUI - https://docs.openwebui.com/tutorials/database/