Inbound Connection Timeout on Port 80 to EC2 Instance (All Software Fails)

Question

Descripción del problema:
"Hola, mi instancia EC2 con la IP Pública Elástica 54.86.157.59 no puede recibir conexiones entrantes en el puerto 80. Cualquier intento de conexión desde internet, incluyendo los servidores de Let's Encrypt, resulta en un error de Timeout during connect.
Hemos realizado un diagnóstico exhaustivo y hemos confirmado que la configuración es correcta. Los pasos realizados son:

DNS: Los registros para edificioburo.cl apuntan correctamente a la IP 54.86.157.59.

AWS Networking:
El Security Group de la instancia permite explícitamente el tráfico HTTP (puerto 80) desde 0.0.0.0/0.
La Network ACL de la subred permite todo el tráfico entrante y saliente.
La Route Table de la subred tiene una ruta a un Internet Gateway.

Configuración de la Instancia:
El firewall del SO (ufw) está inactivo.

Se probó con el servidor web Nginx y luego con Apache. Ambos funcionan perfectamente a nivel interno (curl localhost exitoso), pero las conexiones desde el exterior fallan con el mismo error de Timeout.

Reinicio: La instancia fue reiniciada sin ningún cambio en el comportamiento.

Dado que la configuración de red es correcta y el problema persiste con dos servidores web diferentes, queda demostrado que es un problema a nivel de la plataforma de AWS. Por favor, investiguen por qué el tráfico no llega a la interfaz de red de mi instancia."

re:Post Agent · Accepted Answer

When you're experiencing connection timeouts on port 80 to your EC2 instance despite proper configuration, there are several potential causes to investigate:

1. **Security Group Configuration**: While you've mentioned that your security group allows HTTP traffic on port 80 from 0.0.0.0/0, double-check that there are no conflicting rules or that the rule is properly applied to your instance.

2. **Network Interface**: Verify that the elastic IP is correctly associated with the network interface of your EC2 instance. Sometimes after instance stops/starts, the association might need to be refreshed.

3. **Application Binding**: Ensure your web servers (Nginx/Apache) are binding to all interfaces (0.0.0.0) and not just localhost (127.0.0.1). This would explain why curl localhost works but external connections fail.

4. **Instance Status**: Check if your EC2 instance is experiencing high CPU usage or other performance issues that might cause it to drop connections.

5. **VPC Configuration**: Confirm that your subnet is a public subnet with proper routing to the internet gateway.

6. **Elastic IP Configuration**: If you recently associated or disassociated the Elastic IP, your web server configuration might need to be updated to reflect the new IP address.

7. **Test with Telnet**: Try using telnet from an external machine to test connectivity to port 80 on your instance.

8. **Check Application Logs**: Review your web server logs for any error messages that might indicate why connections are being dropped.

Since you've already verified many of these items, I would recommend:
- Checking if the application is actually listening on the public interface (not just localhost)
- Temporarily disabling any application-level firewalls
- Testing with a simple netcat listener on port 80 to rule out web server configuration issues
- Verifying that the instance's status checks are passing in the EC2 console  
**Sources**  
[EC2 Website Connection Timeout | AWS re:Post](https://repost.aws/questions/QUTXVXoXYVSD2KjSUe-3XFSw/ec2-website-connection-timeout)  
[EC2 Instance that get ERR_CONNECTION_TIMED_OUT | AWS re:Post](https://repost.aws/questions/QUs_ULnAZnQzKC_ECFYkexrw/ec2-instance-that-get-err-connection-timed-out)  
[EC2 Connection issue | AWS re:Post](https://repost.aws/questions/QUkS-pglmQTQSwqmIksb97qw/ec2-connection-issue)  
[Troubleshoot connection errors | AWS re:Post](https://repost.aws/knowledge-center/lightsail-troubleshoot-connection-errors)

Inbound Connection Timeout on Port 80 to EC2 Instance (All Software Fails)

Añade tu respuesta

Contenido relevante