Building the development environment on Amazon Linux 2023 with user_data.txt

Kazuo Yagi — Tue, 25 Apr 2023 14:30:14 +0000

In my case, requirements for an expected development environment on EC2 are as follows.

Requirements

Mandatory:

The following programming language can be available with a specific(latest) version:
- Ruby
- Python
- Node.js
- Rust
Well equipped with build essential and useful tools
- jq
- ripgrep(rg)
- fd(fd-find)
- fzf

Optional:

Could be accessed through EC2 instance connect, the browser-based terminal on AWS Management Console

user_data.txt

https://gist.github.com/kyagi/c28935eeb83b3b03a296a7ed2a179abc#file-user_data-txt

how to run

I use ansible and its collection amazon.aws.ec2_instance collection with my configuration json file with variables to pass over ansible at runtime.

➜ pyenv exec ansible-playbook -i hosts site.yml --extra-vars "@workstation.json"

➜ cat roles/spawn/tasks/main.yml
---
- name: Spawn an ec2 instance
  amazon.aws.ec2_instance:
    name: "{{ instance_name }}"
    key_name: "{{ key_name }}"
    vpc_subnet_id: "{{ vpc_subnet_id }}"
    instance_type: "{{ instance_type }}"
    security_groups: "{{ security_groups }}"
    network:
      assign_public_ip: true
      private_ip_address: "{{ private_ip_address }}"
    image_id: "{{ image_id }}"
    user_data: "{{ lookup('file', 'user_data.txt') }}"
    volumes:
      - device_name: /dev/xvda
        ebs:
          volume_size: 30
          volume_type: gp2
          delete_on_termination: true
    tags:
      Owner: "{{ owner }}"
      Stack: "{{ stack }}"
      Project: "{{ project }}"
      Application: "{{ application }}"
      Ec2InstanceConnectAccess: "{{ ec2_instance_connect_access }}"

➜ cat workstation/workstation.json
{
  "instance_name": "workstation",
  "key_name": "__key-name__",
  "vpc_subnet_id": "__subnet-id__",
  "instance_type": "t4g.small",
  "security_groups": ["__sg-id1__", "__sg-id2__"],
  "private_ip_address": "10.0.0.10",
  "image_id": "resolve:ssm:/aws/service/ami-amazon-linux-latest/al2023-ami-kernel-default-arm64",
  "owner": "kyagi",
  "stack": "deveplopment",
  "project": "workstation",
  "application": "hodgepodge of development tools",
  "ec2_instance_connect_access": "true"
}

Shamshir: Github Actions to attach a label to pull requests based on the number of approvals

Kazuo Yagi — Sun, 06 Feb 2022 18:07:27 +0000

Introduction

Shamshir is a Github Action to attach a label to pull requests based on the number of approvals.

You would find it useful if your team's git repository is run under some conventions that requires 2 or more approvals from members to let a pull request merged or release.

Shamshir will make it easy to find out which pull request can be "releasable" by attaching a label to each pull request based on the number of approvals.

It is designed to be run both on stand-alone server and on Github Actions.

https://github.com/kyagi/shamshir

Forem: Kazuo Yagi