edh-stats/QUICK_FIX.md

Quick Fix for EDH Stats Deployment Issues

Issue 1: Backend - "unable to open database file"

Issue 2: Frontend - "host not found in upstream 'backend'"

This guide fixes both issues immediately.

Step 1: Stop Services

docker-compose down

Step 2: Update nginx.prod.conf

The nginx config needs to dynamically resolve the backend hostname. On your server, edit the nginx config:

Replace the HTTP/HTTPS server blocks with this simpler version:

# Use resolver to dynamically resolve backend hostname
resolver 127.0.0.11 valid=10s;
set $backend_backend "backend:3000";

server {
    listen 80;
    server_name _;
    root /usr/share/nginx/html;
    index index.html;

    # Rate limited API proxy
    location /api/ {
        limit_req zone=api burst=20 nodelay;
        proxy_pass http://$backend_backend;
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header 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_cache_bypass $http_upgrade;
        proxy_read_timeout 86400;
    }

    # Rate limited static files
    location ~* \.(js|css|png|jpg|jpeg|gif|ico|svg)$ {
        limit_req zone=static burst=50 nodelay;
        expires 1y;
        add_header Cache-Control "public, immutable";
    }

    # Main application routes
    location / {
        limit_req zone=static burst=10 nodelay;
        try_files $uri $uri/ /index.html;
    }

    # Health check endpoint
    location /health {
        access_log off;
        return 200 "healthy\n";
        add_header Content-Type text/plain;
    }

    # Error pages
    error_page 404 /index.html;
    error_page 500 502 503 504 /50x.html;
    location = /50x.html {
        root /usr/share/nginx/html;
    }
}

Step 3: Update docker-compose.yml - Frontend Ports

If other containers are using ports 80 and 443, update the frontend service to use alternate ports:

  frontend:
    image: ghcr.io/zlorfi/edh-stats-frontend:v1.0.5
    ports:
      - '38080:80'  # Changed from 80:80
      - '30443:443' # Changed from 443:443
    restart: unless-stopped
    networks:
      - edh-stats-network

Then access the app at:

• HTTP: http://your-server:38080
• HTTPS: https://your-server:30443 (when SSL configured)

Step 4: Update docker-compose.yml - Init Service

Update the init-db service to use more permissive permissions:

  init-db:
    image: alpine:latest
    volumes:
      - sqlite_data:/app/database/data
      - app_logs:/app/logs
    command: sh -c "mkdir -p /app/database/data /app/logs && chmod 777 /app/database/data /app/logs && touch /app/database/data/.initialized && echo 'Database directories initialized'"
    networks:
      - edh-stats-network
    restart: "no"

Key changes:

• Changed chmod 755 to chmod 777 for full read/write permissions
• Added touch /app/database/data/.initialized to create a test file
• Added restart: "no" so the init service doesn't restart

Step 5: Clear Existing Volumes (Fresh Start)

⚠️ WARNING: This will DELETE your database if you have data!

If you want to start fresh:

docker volume rm edh-stats_sqlite_data edh-stats_app_logs 2>/dev/null || true

Step 6: Start Services

docker-compose up -d

Step 7: Monitor Startup

# Watch the init-db service create directories
docker-compose logs -f init-db

# Then watch the backend startup
docker-compose logs -f backend

# Check if all services are running
docker-compose ps

Expected Output

After init-db runs and exits, you should see:

backend-1   | Database initialized successfully
backend-1   | Server started on port 3000

And:

frontend-1  | ... ready for start up

Verify Everything Works

# Check API is responding (backend runs on port 3000 internally)
curl http://localhost:3000/api/health

# Check frontend is responding (now on port 38080)
curl http://localhost:38080/

# All containers running?
docker-compose ps
# Should show: backend (running), frontend (running), init-db (exited)

# Access the app in your browser
# HTTP: http://your-server-ip:38080
# HTTPS: https://your-server-ip:30443 (when SSL is configured)

What Changed and Why

Nginx Config Issue

The old config tried to resolve backend hostname at startup time before the backend container was ready. By using resolver and set $backend_backend, nginx defers resolution until request time, when the backend is running.

Database Permissions Issue

Docker volumes sometimes have restricted permissions. By changing to chmod 777, the container has full permission to create and write the database file. The touch command creates a test file to verify permissions work.

Init Service

The init-db service runs BEFORE the backend and ensures:

1. Directories exist
2. Directories are writable
3. Backend can then create the database successfully

Still Having Issues?

Check volume exists:

docker volume ls | grep edh-stats

Check volume path is writable:

VOLUME_PATH=$(docker volume inspect edh-stats_sqlite_data | grep -o '"Mountpoint": "[^"]*' | cut -d'"' -f4)
echo "Volume at: $VOLUME_PATH"
ls -la "$VOLUME_PATH"

Remove and recreate volumes:

docker-compose down
docker volume rm edh-stats_sqlite_data edh-stats_app_logs
docker-compose up -d

Check logs for specific errors:

docker-compose logs --tail 100 backend
docker-compose logs --tail 100 frontend