chat/nginx.hanzo.conf

user nginx;
worker_processes auto;
error_log /var/log/nginx/error.log warn;
pid /var/run/nginx.pid;

events {
    worker_connections 1024;
}

http {
    include /etc/nginx/mime.types;
    default_type application/octet-stream;
    
    log_format main '$remote_addr - $remote_user [$time_local] "$request" '
                    '$status $body_bytes_sent "$http_referer" '
                    '"$http_user_agent" "$http_x_forwarded_for"';
    
    access_log /var/log/nginx/access.log main;
    
    sendfile on;
    tcp_nopush on;
    tcp_nodelay on;
    keepalive_timeout 65;
    types_hash_max_size 2048;
    
    # Gzip compression
    gzip on;
    gzip_vary on;
    gzip_min_length 10240;
    gzip_proxied expired no-cache no-store private auth;
    gzip_types text/plain text/css text/xml text/javascript application/x-javascript application/xml application/javascript application/json;
    gzip_disable "MSIE [1-6]\.";
    
    # Security headers
    add_header X-Frame-Options "SAMEORIGIN" always;
    add_header X-Content-Type-Options "nosniff" always;
    add_header X-XSS-Protection "1; mode=block" always;
    add_header Referrer-Policy "no-referrer-when-downgrade" always;
    
    # Upstream definitions
    upstream chat_backend {
        server chat:3080;
    }
    
    upstream router_backend {
        server router:4000;
    }
    
    # Rate limiting
    limit_req_zone $binary_remote_addr zone=general:10m rate=10r/s;
    limit_req_zone $binary_remote_addr zone=api:10m rate=30r/s;
    
    # Main server block
    server {
        listen 80;
        server_name localhost;
        
        # Health check endpoint
        location /health {
            access_log off;
            return 200 "healthy\n";
            add_header Content-Type text/plain;
        }
        
        # Chat application
        location / {
            limit_req zone=general burst=20 nodelay;
            
            proxy_pass http://chat_backend;
            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;
            
            # WebSocket support
            proxy_read_timeout 86400;
        }
        
        # Chat application REST API (namespaced /v1/chat/* -> chat backend).
        # MUST precede location /v1/ so the OpenAI-compat inference proxy
        # (router_backend) never shadows the app surface. Longest-prefix match
        # also guarantees precedence, but we keep it explicit and first.
        location /v1/chat/ {
            limit_req zone=api burst=50 nodelay;

            proxy_pass http://chat_backend;
            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;

            # SSE / streaming (chat completions stream via the agents route)
            proxy_buffering off;
            proxy_cache off;
            chunked_transfer_encoding on;
            proxy_read_timeout 86400;
        }

        # Hanzo Router API Gateway (OpenAI-compat inference: /v1/models, /v1/embeddings, ...)
        location /v1/ {
            limit_req zone=api burst=50 nodelay;

            proxy_pass http://router_backend;
            proxy_http_version 1.1;
            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;
            
            # Streaming support
            proxy_buffering off;
            proxy_cache off;
            proxy_set_header Connection '';
            chunked_transfer_encoding on;
            
            # Timeouts for long-running LLM requests
            proxy_connect_timeout 300s;
            proxy_send_timeout 300s;
            proxy_read_timeout 300s;
        }
        
        # Hanzo Router Admin UI
        location /ui {
            proxy_pass http://router_backend;
            proxy_http_version 1.1;
            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;
        }
        
        # Static assets with caching
        location ~* \.(jpg|jpeg|png|gif|ico|css|js|svg|woff|woff2|ttf|eot)$ {
            proxy_pass http://chat_backend;
            expires 1y;
            add_header Cache-Control "public, immutable";
        }
    }
    
    # HTTPS server block (uncomment when SSL certificates are available)
    # server {
    #     listen 443 ssl http2;
    #     server_name localhost;
    #     
    #     ssl_certificate /etc/nginx/ssl/cert.pem;
    #     ssl_certificate_key /etc/nginx/ssl/key.pem;
    #     ssl_protocols TLSv1.2 TLSv1.3;
    #     ssl_ciphers ECDHE-RSA-AES128-GCM-SHA256:ECDHE:ECDH:AES:HIGH:!NULL:!aNULL:!MD5:!ADH:!RC4;
    #     ssl_prefer_server_ciphers on;
    #     
    #     # Include all location blocks from above
    # }
}