Immutable-Infrastructure

BACKGROUND INFORMATION

Before Virtualization and Cloud Computing:

• Server infrastructure was enpasulated in physical servers that were:

  • expensive
  • time consuming to create
  • setup took days or weeks because of the need to order new hardware, configuration the machine, and other manual processes.

• Mutable Infrastructure = older way of confguring infrastructure that was leveraged prior to the emergence of virtualization and cloud computing that has allowed for the shift to immutable infrastructure

Shift to Immutable Infrastructure

With Immutable Infrastructure, we leverage virtualization and cloud computing to allow for new deployment workflows in which new servers can be provisioned quickly and programmatically

• virtualization = when a simulated computing environment is created rather than a physical environment
• cloud computing = using network of remote servers hosted via internet to store, manage + process data

Benefits of Cloud Computing

1. No Configuration Drift or Snowflake Serveres
   • config. drift = changes to software or hardware are mad ein data center environment and not systematically tracked
   • snowflake servers = a server that requires special configuration beyond the means of automated deployment (servers become difficult for replication because each snowflake server has a unique fragile part of overall infrastructure)
2. Consistent staging environments and easy horizontal scaling
3. Simple rollback and recovery processes

REPOSITORY CONTENTS

images:

• directory storing packer image template (image.pkr.hcl)

instances:

• main.tf : declaration of terraform resources, aws_instance resource, and output public ip address
• terraform.tfvars : region declaration
• variables.tf : terraform vars. w/ associated descriptions

scripts:

• setup.sh

Additional files:

• tf-packer : stores private key
• tf-packer.pub : contains ssh key

NOTE: This repository will be associated with an Instruqt lab tracK. The track will utilize this repo to show the process of deploying a machine image, updating the image, and showing it redeploy

NOTE: This is just a copy from the Companion code repository. Using the same template to get a feeling of building an image with packer and deploying it using terraform. There is a lot of different ways to build an image and use different provisioners for web-apps: I found this the most easiest way instead of using a .service file.. We just use a shell script and issue a go-get with a github link to your web-app.

Resources:

1. https://www.digitalocean.com/community/tutorials/what-is-immutable-infrastructure
2. https://www.sumologic.com/insight/mutable-immutable-infrastructure/
3. https://www.hashicorp.com/resources/what-is-mutable-vs-immutable-infrastructure
4. https://www.hashicorp.com/resources/modern-infrastructure-automation-with-packer-terraform-and-consul

Web-App:

1. Make sure to set ssh-rsa keys: tf-packer and tf-packer.pub
2. Also when ssh-ing to server the directory path to run the web-app is:
3. /go/pkg/mod/github.com/sabinlehaci/go-web-app@v0.0.0-20220706130610-b856cb378344$