Add bulletproof startup script to fix django_site migration issues

- Create startup.sh with proper step-by-step setup and clear logging
- Explicit sites migration before configure_site command
- Better error handling with set -e
- Clean command structure in docker-compose
- This should resolve the 'relation django_site does not exist' error

The startup script runs migrations in correct order and provides clear feedback.

docker-compose.dokploy-simple.yml

@@ -17,6 +17,8 @@ services:
       - db
     networks:
       - dokploy-network
+    command: >
+      sh -c "chmod +x /app/startup.sh && /app/startup.sh"
 
   db:
     image: postgres:15-alpine
@@ -35,10 +37,3 @@ volumes:
 networks:
   dokploy-network:
     external: true
-    command: >
-      sh -c "python manage.py migrate &&
-             python manage.py create_groups &&
-             python manage.py setup_social_apps &&
-             python manage.py configure_site --domain=dt.netcoptech.com --name='Django Template' &&
-             python manage.py collectstatic --noinput &&
-             gunicorn --bind 0.0.0.0:8000 --workers 3 --timeout 120 django_project.wsgi:application"

startup.sh

@@ -0,0 +1,31 @@
+#!/bin/bash
+
+set -e
+
+echo "🚀 Starting Django application setup..."
+
+echo "⏳ Waiting for database..."
+sleep 5
+
+echo "📦 Running database migrations..."
+python manage.py migrate --noinput
+
+echo "🌐 Ensuring Sites framework is set up..."
+python manage.py migrate sites --noinput
+
+echo "👥 Creating user groups..."
+python manage.py create_groups
+
+echo "🔗 Setting up social applications..."
+python manage.py setup_social_apps
+
+echo "🏠 Configuring site domain..."
+python manage.py configure_site --domain=dt.netcoptech.com --name="Django Template"
+
+echo "📁 Collecting static files..."
+python manage.py collectstatic --noinput
+
+echo "✅ Setup completed successfully!"
+
+echo "🚀 Starting Gunicorn server..."
+exec gunicorn --bind 0.0.0.0:8000 --workers 3 --timeout 120 django_project.wsgi:application