Loadbalancing SFTP with TCP connection timed out

Question

I have a sftp process running on a machine on Amazon EC2. Connecting with Filezilla works flawlessly.

I tried to create an Amazon load balancer over TCP:TCP for this process, internet facing, but I can only get

Connecting to XXX.elb.amazonaws.com...

Response: fzSftp started, protocol_version=6

Command: open "XXX.elb.amazonaws.com" 22

Error: Network error: Connection timed out

Error: Could not connect to server

Status: Waiting to retry...

Status: Connecting to XXX.elb.amazonaws.com...

Response: fzSftp started, protocol_version=6

Command: open "XXX.elb.amazonaws.com" 22

Error: Network error: Connection timed out

Error: Could not connect to server

when trying to connect with Filezilla. It seems that I can't connect through the TCP loadbalancer ( I double checked the ports used and the health check is passing: 'InService' ). How can I make it work?

Answer 1

Just found out that the security group I used did not authorize the use of the specified ports.

The solution is consequently to create a new security group with the right ports allowed, or open new ports on the current security group.