Route ports to instances using Amazon VPC - amazon-web-services

Question

I have a simple VPC on Amazon EC2 with two instances: an SFTP server, and a web server. I want to route incoming traffic from the internet on port 22 to the SFTP server and ports 80 and 443 to the web server.
I'm having trouble finding exactly how to do this. It's extremely simple with all hardware routers I've ever used.
Can anyone point me to documentation/examples/or just tell me how to do it?
Edit:
I wasn't clear that I want remote hosts to be able to make a request to a single IP address, but on different ports, and have different ports routed to different instances; I already have my security groups set so I can accept traffic to the public IP addresses of each instance.

Answer 1

Below are the steps you will need to follow to achieve your use case.
As you said both servers are in VPC you need to make sure they are
in a Public Subnet with a internet gateway attached to it.(This can
be achieved in Private subnet too but will become more complex
process for you) For more information check this
https://docs.aws.amazon.com/AmazonVPC/latest/UserGuide/VPC_Scenario1.html
EC2 Server 1 for SFTP you need to open SSH port 22 in the security group by adding your IP address or open to world rule. Then you can SFTP using your preferred SFTP client.
EC2 Server 2 for port 80 and 443. you need to add two rules in the security group for your instance. Check below screenshot. Once that is done you will be able to route http traffic through port 80 and 443.
More information on security groups is here.
https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-network-security.html
About your Edit. You have to run a instance in front of both the instance.
Add a HTTP proxy to this new created instance. And then the port redirect should be achieved using IP Tables. Check this below IP tables config you need to add to that server. Let say your HTTP proxy server is Server 1. Port 22 server is Server 2 and Port 80 and 443 Server is Server 3. So Server 1 IP tables will have below rules.
#!/bin/sh
echo 1 > /proc/sys/net/ipv4/ip_forward
iptables -F
iptables -t nat -F
iptables -X
iptables -t nat -A PREROUTING -p tcp --dport 80 -j DNAT --to-destination <Server3-IP>:80
iptables -t nat -A POSTROUTING -p tcp -d <Server3-IP> --dport 80 -j SNAT --to-source <Server1-IP>
iptables -t nat -A PREROUTING -p tcp --dport 22 -j DNAT --to-destination <Server2-IP>:80
iptables -t nat -A POSTROUTING -p tcp -d <Server2-IP> --dport 22 -j SNAT --to-source <Server1-IP>
iptables -t nat -A PREROUTING -p tcp --dport 443 -j DNAT --to-destination <Server3-IP>:80
iptables -t nat -A POSTROUTING -p tcp -d <Server3-IP> --dport 443 -j SNAT --to-source <Server1-IP>

Answer 2

As of today, Amazon Elastic Load Balancers have a new "Application Load Balancer" mode which supports routing requests to different EC2 instances based on things such as the HTTP path, and port.
To set this up, do the following:
Create a new ELB, using the new "Application Load Balancer" mode.
Create one target group for port 22.
Add your EC2 instances that will be the SSH targets to this target group.
Create another target group for ports 80 and 443.
Add your EC2 instances that will be the HTTP/HTTPs targets to this target group.
Using this system, SSH requests to the ELB will be forwarded to the SSH instance, and HTTP/HTTPs requests to the ELB will be forwarded to the web servers