Creating personal VPN service has never been so easy. (c) said no one so far

Seriously though, using this Terraform project you'll be able to easily create your own ready-to-use IPsec VPN along with Telegram MTProto proxy.
Combining Terraform and Ansible it automatically deploys and provisions AWS EC2 instance with all VPN-needed services:

• IPsec VPN based on strongSwan
• official Docker image used to run Telegram Messenger MTProto proxy

Default IPsec configuration within this script is suitable for two cases:

• site-to-site VPN
• remote access VPN
  

Make sure Terraform version 0.12 or higher is installed.
Make sure Ansible version 2.5 or higher is installed.
Perform terraform init command within repository directory.

All variables are located within variables.json file and are being used both by Terraform and Ansible during deployment and provisioning processes.
Variables are splitted into several sections and need to be changed by your values:

• instance - specify AWS region, AMI and instance type; current solution was developed and is only tested with Centos 7 operating system
• connections - specify AMI username, SSH port you'd like to use (Terraform will change it during initial deployment), service ports and ssh-keys
• provisioner
  • tunnel - set peer_wan to your routers public WAN IP and peer_lan to your LAN network (if you want to use IPsec site-to-site VPN)
  • ipsec - user and shared_key are credentials for remote access VPN, psk is also needed for connection establishing
  • telegram - set secret that will be specified within your Telegram configuration Leave the route53 section with defaults if you don't have a domain parked on AWS or don't want to bind a domain to the created instance. Otherwise, set enabled option to true and specify your domain and AWS route53 zone id (should be created manually before).
    Also, environment variables containing AWS access and secret keys should be created:

export AWS_ACCESS_KEY="qwerty"
export AWS_SECRET_KEY="qwerty"

Simply run ./pipeline.sh file and type yes when terraform asks about deploy.
It will generate a new SSH key pair, deploy all AWS resources required and finally spin up AWS instance followed by software provisioning.

Starting a pipeline...
==============================================================
Generating a key-pair
-----------------------------------
Generating public/private rsa key pair.
centoskey already exists.
Overwrite (y/n)? Your identification has been saved in centoskey.
Your public key has been saved in centoskey.pub.
The key fingerprint is:
SHA256:xabDOaqA14tTigfHpYtRbg6240gM/ZVwGb6uA1/gWLY
The key's randomart image is:
+---[RSA 2048]----+
|      .          |
|     . o .       |
|    . +   +      |
| .. =o + =       |
|.+.B o+ S        |
|*oOoEo.. o       |
|o%+*o.o          |
|=+*oo+           |
|ooo.+.           |
+----[SHA256]-----+
==============================================================
Applying terraform configuration
-----------------------------------

Deployment output will contain public IP address (that is binded to the specified domain if configured).

Apply complete! Resources: 4 added, 0 changed, 0 destroyed.

Outputs:

public_ip = 13.13.0.1
==============================================================
Done!
==============================================================

You can use specified domain name or a public IP of the deployed instance as a VPN server destination and Telegram proxy server address.

There is a Jenkinsfile under the jenkins-pipeline folder. It can be used for the infrastructure deployment and provisioning.