Context

Telegram bots offer a fast and efficient way to prototype applications without requiring custom UI or authentication. They can be deployed on a virtual machine to maintain state or in a serverless environment. A serverless Telegram bot hosted on AWS Lambda benefits from 1 million free compute requests and supports concurrent execution, unlike stateful bots on VMs with limited compute capacity. However, configuring a serverless setup can be time-consuming.

This template simplifies the deployment of a serverless Telegram bot, providing a solid foundation for users familiar with Infrastructure as Code (IaC) and AWS. It also serves as a quick-start guide for those looking to deploy a simple Telegram bot. However, extensive customization may require additional onboarding to gain familiarity with Terraform.

Architectural Overview

Pre-requisites

1. Terrform - refer to installation guide.
2. AWS Account credentials see AWS Build infrastructure tutorial.
   • AWS_ACCESS_KEY_ID
   • AWS_SECRET_ACCESS_KEY
3. Create an AWS Bucket to store terraform state file (Note the bucket name)
4. TELEGRAM_BOT_TOKEN - Generated from bot father see section on obtaining bot token.
5. TELEGRAM_BOT_NAME - The name of your telegram bot to be used for naming AWS infrastructure.

Getting started

This section explains how to get from an AWS token and telegram the ensure pre-requisites are met

Initialise Python Virtual Environment and requirements.txt

The requirements.txt file will be referenced during infrastructure deployment to ensure required dependencies are present in lambda layers. Navigate to source directory, install and activate virtual environment cd source pip3 -m venv venv source venv/bin/activate

Install initial dependencies and any additional dependencies into venv
`pip3 install -r requirements.txt`
`pip3 install <your dependencies>`

Export all dependencies requirements.txt
`pip3 freeze > requirement.txt`

Deploy telegram bot via terraform

1. Export AWS credentials to terminal
2. Replace state file reference in main.tf
3. Create a new file by name terraform.tfvars and populate variables bot_token and bot_name. See terraform.tfvars.example for reference.
4. terraform init and terraform apply. Optionally run terraform validate and terraform plan to verify config before apply.

Developer guide

This section explains the elements of the template to aid in providing context required to perform customisations.

Main.tf

Terrform provider configuration with remote state file definitions.

Scripts

prepare_layers.sh - bash script to install AWS Linux compatible python depedencies for lambda layers register_telebot.sh - bash script to set webhook enabling telegram server to foward messaged send to our bot to our API Gateway

terraform_data.tf

Invokes the scripts in above section on scripts based on lifecycle triggers. Only runs layers script if there is a change in requirements.txt

apigw.tf

API gateway paths configured to invoke lambda function.

lambda.tf

Lambda layers, function and role configuration to execute source code. Update lambda role if you plan to integrate lambda other AWS services (eg. DynamoDB).

data.tf

Script to package source code and dependencies requireed for lambda source and layers. Source code packaging excludes venv folder. Update file if your virtual environment is of a different name.