EC2 Instance Starter

AWS EC2 Instance Initialization and Secure Connection via AWS SSM

This repository contains a Python script for automating the initialization of Amazon EC2 spot instances, enabling efficient access to savings of up to 80% compared to on-demand prices. The script leverages the boto3 library to launch instances from a pre-existing Amazon Machine Image (AMI) using a user-defined launch template. It's particularly useful for setting up private generative AI inference and training Stable Diffusion via ComfyUI, A1111, as well as LLMs.

Key Features

• Secure Connection via AWS Systems Manager (SSM): The script sets up a secure connection using AWS SSM, eliminating the need for traditional SSH keys and AWS security groups. This feature enhances security and simplifies access management.

• Port Forwarding Configuration: The script can configure port forwarding, allowing seamless interaction with AI servers as if they were local. This feature is especially beneficial for developers needing a local-like environment for inference. Default port in the config document is 8188, which is the ComfyUI default. eg. http://127.0.0.1:8188 would reach your EC2 instance, but without needing to expose any ports to the internet in AWS Security Groups.

• Flexible Network Design: The script supports the creation of a highly secure instance within a private Virtual Private Cloud (VPC) without an internet gateway using AWS PrivateLink. However, this is optional, and users can choose to have their instance within a VPC that includes an internet gateway if they prefer, without needing to expose any ports to the internet.

• Automated Updates for User-Defined Applications: The config.yaml file includes a user_data section that is executed upon EC2 instance launch. This feature enables automatic updates for repositories, extensions, and ComfyUI custom nodes, among other elements.

Getting Started

These instructions will guide you on how to use this script for starting and connecting to an EC2 instance.

Prerequisites

Before you start, make sure you have:

• An AWS account
• AWS CLI installed and configured
• Install AWS System Manager plug-in
• A predefined EC2 launch template
• An AMI you can use to launch, which already has SSM Agent
• Ensure your launch template will attach to the instance an IAM role with SSM permissions
• Appropriate IAM permissions

Installation

1. Clone the repo

git clone https://github.com/cavit99/Start-EC2.git

2. Install the required packages

pip install -r requirements.txt

Usage

Run the script

python start_ec2.py

Follow the prompts to select your launch template and start the instance.

Threading Logic

This script uses Python's threading module to handle the output of the SSM port forwarding session in real-time.

1. Main Thread: This is the primary thread that runs the main function and controls the overall flow of the script.

2. Secondary Thread: This thread is created to handle the output of the SSM port forwarding session in real-time. It runs a function called handle_output, which reads the output line by line and logs it. It also sets an event when the port forwarding session is ready.

Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

Distributed under the MIT License. See LICENSE for more information.

Contact

Cavit Erginsoy - cavit@erginsoy.com

Project Link: https://github.com/cavit99/Start-EC2