Configure EC2 Instance

Scenarios:

1. Ensure the EC2 machine created has
   • some softwares (like git etc ) instance
   • services like apache/tableu etc are started
   • some files downloaded from internet
   • Solution:
     • for software installations we might create an ami (but ami requires storage cost)
     • Services can be automatically started if they are configured in that way in AMI otherwise change AMI to meet your needs
     • If we write some kind of startup script, which gets executed when the machine is created, then it will help.
2. Lets install apache server on amazon linux instance
   • As shown in the image below create an instance by changing user-data

Cloud-init

• A Startup script which gets launched when the vm is created and this approach is adopted by multiple clouds using a standard called as cloud-init.

• In Cloud-init, user is expected to write a yaml file as mentioned over here

• For overall documentation of cloud-init refer here

• Userdata is stored in a service running in the ec2 machine on the ipaddress ‘169.254.169.254’

  • To get latest userdata

  curl http://169.254.169.254/latest/user-data
  

  • To run user-data without restarting the machine

  sudo -i
  curl http://169.254.169.254/latest/user-data > userdata.sh
  sh ./userdata.sh
  

  • The logs of the user data are stored in /var/log/cloud-init-output.log

• Instance-Metadata: Refer Here

• Refer Here for simple cloud-init

• simple example of cloud-init

#cloud-config
repo_update: true
repo_upgrades: all

packages:
  - git
  - tree

runcmd:
  - [echo, 'helloworld']
  - [echo, 'helloworld1']