Introduction
Installing caddy
Installing caddy
Configuring caddy
Accessing logs
Using tcpdump to inspect http/3 protocol
Using jq to make logs readable as csv
Final note

Introduction

In this article we will install Caddy web server, in order to make use of the http/3 protocol on the hosted web-site.
Caddy will serve web-pages using the older http/1.1 protocol for older web browsers.
The server will also automatically renew the SSL web-site certificate and, when using http/3, browsers will use TLS 1.3, a secure protocol, and UDP, instead of TCP; this reduces latency and saves bandwidth.

In this article, we will deploy the same AWS infrastructure as in the Github article.
We will use the same main.tf file, we’ve changed the name of the .tfstate terraform state file in the S3 bucket.

The .gitlab-ci.yml file in the root of the project contains the 3 stages, plan/apply and destroy. The plan/apply stages are triggered by any changes committed to the main.tf terraform file.
They can also be run manually, from the Build/pipelines section, where 3 checkmarks will appear under the Stages header, for the pipeline.
The destroy stage can be invoked manually by clicking on the last checkmark and on the run button at the end.
The AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY variables need to be added manually to the Settings > CI/CD section.

Introduction

In this article we will use Github Actions to deploy terraform infrastructure.
An s3 bucket will be used to save and retrieve the terraform state file, in case we need to remove infrastructure via terraform destroy.
For the bucket, the AWS account will need to have the permissions below (attached to the bucket via an inline policy).

Steps to Set up GitHub Actions for Terraform:

1. Create a GitHub Repository: