fix: move cron job to dedicated compose service, avoids LXC runc PID errors

docker-compose.yml

@@ -46,6 +46,37 @@ services:
     volumes:
       - static_files:/app/staticfiles
 
+  # Runs the nightly coordinator status check (django-crontab job)
+  # Avoids needing a cron daemon inside the web container, which causes
+  # runc PID errors in unprivileged LXC environments.
+  cron:
+    build: .
+    restart: unless-stopped
+    entrypoint: >
+      sh -c "echo '0 6 * * * python /app/manage.py nightly_check_active_status >> /proc/1/fd/1 2>&1'
+      | crontab - && crond -f -d 8"
+    depends_on:
+      db:
+        condition: service_healthy
+    environment:
+      SECRET_KEY: ${SECRET_KEY}
+      DEBUG: ${DEBUG:-False}
+      ALLOWED_HOSTS: ${ALLOWED_HOSTS:-localhost}
+      SQL_ENGINE: django.db.backends.postgresql
+      SQL_DATABASE: ${SQL_DATABASE:-busmanager}
+      SQL_USER: ${SQL_USER:-busmanager}
+      SQL_PASSWORD: ${SQL_PASSWORD}
+      SQL_HOST: db
+      SQL_PORT: 5432
+      EMAIL_HOST: ${EMAIL_HOST:-}
+      EMAIL_PORT: ${EMAIL_PORT:-25}
+      AZURE_CLIENT_ID: ${AZURE_CLIENT_ID:-}
+      AZURE_CLIENT_SECRET: ${AZURE_CLIENT_SECRET:-}
+      AZURE_REDIRECT_URI: ${AZURE_REDIRECT_URI:-}
+      AZURE_AUTHORITY: ${AZURE_AUTHORITY:-}
+      TELSTRA_CLIENT_ID: ${TELSTRA_CLIENT_ID:-}
+      TELSTRA_CLIENT_SECRET: ${TELSTRA_CLIENT_SECRET:-}
+
 volumes:
   postgres_data:
   static_files: