Anonymous Telegram Chat Bot

Deployment

Deploy Init Resources

Now we create and deploy the resources needed for the main stack including a S3 bucket to store Terraform state, a DynamoDB table to persist Terraform state lock and a S3 bucket to deploy Lambda function code bundles. Go to infra/init directory and create a file named terraform.tfvars:

init_aws_region        = "<AWS_REGION>"
terraform_state_bucket = "<S3_TERRAFORM_STATE_BUCKET_NAME>"
init_lambda_bucket     = "<S3_LAMBDA_CODE_BUCKET>"
github_repo            = "<GITHUB_PROFILE/REPO_NAME>"

Note: The Terraform state for this init stack is being kept locally.

Now plan and apply the changeset:

terraform init # Run once
AWS_PROFILE=<AWS_PROFILE> terraform plan -out init.tfplan
AWS_PROFILE=<AWS_PROFILE> terraform apply "init.tfplan"

Build and Bundle

cd bot
``
GOARCH=amd64 GOOS=linux go build -o bootstrap main.go

Note: In order to use the binary as the Lambda handler, it should be named bootstrap.

Create a zip bundle from the built binary

zip lambda_function.zip bootstrap

Upload Bundle

Upload the build zip bundle to S3:

aws s3 cp lambda_function.zip s3://<S3_LAMBDA_CODE_BUCKET>/lambda_function.zip --profile <AWS_PROFILE>

Deploy Main AWS Resources

Add Terraform Variables File

Create a file named infra/terraform.tfvars:

aws_region      = "<AWS_REGION>"
lambda_bucket   = "<S3_LAMBDA_CODE_BUCKET>"
bot_token       = "<TELEGRAM_BOT_TOKEN>"

Initialize Terraform

Create a Terraform backend configuration file named infra/backend_config.hcl:

region       = "<AWS_REGION>"
bucket       = "<S3_TERRAFORM_STATE_BUCKET_NAME>"

Initialize the main Terraform stack:

cd infra

AWS_PROFILE=<AWS_PROFILE> terraform init -backend-config backend_config.hcl # Run once

Deploy

Use terraform plan to create a changeset and run terraform apply command to deploy the changeset on AWS:

cd infra

AWS_PROFILE=<AWS_PROFILE> terraform plan -out main.tfplan
AWS_PROFILE=<AWS_PROFILE> terraform apply "main.tfplan"

Register Bot Webhook

Get the API_GATEWAY_STAGE_INVOCATION_URL from terraform output and register it as the webhook on telegram.

curl https://api.telegram.org/bot<BOT_TOKEN>/setWebhook \
-F "url=<API_GATEWAY_STAGE_INVOCATION_URL>

Local Development

Use docker compose to run local DynamoDB and also a local development web server on port 8080:

export BOT_TOKEN=<BOT_TOKEN>
docker compose up -d

After applying a change in the bot code, to update the local bot in the docker container:

docker compose restart bot

Run a reverse proxy tool with a public secure API gateway on your local 8080 port.
For example using ngrok:

ngrok http 8080

Register the local bot on Telegram using the ngrok's forwarding endpoint:

curl https://api.telegram.org/bot<BOT_TOKEN>/setWebhook \
-F "url=<NGROK_FORWARDING_ENDPOINT>

You can see the incoming update requests sent by Telegram on ngrok's web interface: http://127.0.0.1:4040.