Skip to content

Latest commit

 

History

History
47 lines (46 loc) · 3.14 KB

devops-hands-on-path.md

File metadata and controls

47 lines (46 loc) · 3.14 KB
Raw

Step By Step

  • Install Ubuntu on your computer.
  • Install Virtualbox
  • Install a Ubuntu server inside virtualbox.
  • Install Apache, PHP, MySQL inside the server
  • Install WordPress on the server and make it work.
  • Now add Nginx as a reverse proxy in front of it.
  • Now create another server and install just MySQL on it.
  • Dump the MySQL database from the first server and restore it on the new one.
  • On the old server configure WordPress to use the database on the new server instead of the database on the old one.
  • Uninstall the MySQL from the old server.
  • Now create another server
  • Install Nginx on it.
  • Configure this new server to be a reverse proxy to the old one (pointing to Apache).
  • Now remove Nginx from the old Server.
  • Now install another server.
  • Install NFS on it.
  • Mount the NFS on the old server.
  • Move your WordPress files to the NFS volume on the old server.
  • Make Apache read the Files on the NFS volume.
  • Now install another server
  • Install Apache + PHP on it.
  • Mount the NFS server and make Apache read the Files from NFS.
  • Check to see if WordPress is working on this server.
  • Now configure Nginx to be a reverse proxy to both Apache severs. Congratulations, you put up a full, web HA enabled, escalable, web stack.
  • Now, re-do the above using Ansible to configure the servers, fetch the WordPress codebase from GitHub and restoring the database dump (Hint: you can keep the database dump for now together with your Ansible files). Congratulations, you made your first system using configuration management.
  • Now, do the same using Ansible to provision AND configure the servers. Congratulations, now you know how to provision AND configure a server stack using configuration management system automation.
  • Now, refactor your Ansible automation to do the same, but on provision on AWS instead of Virtualbox. Congratulations, your first cloud based web-stack.
  • Now refactor your Ansible automation codebase to do the same but restoring the database dump from S3. Congratulations, now you know how to restore dumps from a reliable, world available, object storage.
  • Now, refactor your Ansible automation to do the same but using EFS instead of your NFS server. Congratulations, your first Ansible provisioning using a AWS resource.
  • Now refactor your Ansible codebase to do the same but using ELBs instead of the Nginx reverse proxy. Congratulations, now you removed one single point of failure from your web-stack.
  • Now refactor your Ansible codebase to do the same but replacing the MySQL database server by a RDS. Congratulations, now you removed the remaining bottleneck of your web-stack.
  • Now refactor your Ansible automation codebase to do the same but using Packer generated Golden Images + AutoScale Groups + Launch Configuration. Congratulations, now you have a self-healing, auto-scale enabled web-stack. If you reach this point, let me know... We may be hiring! 😊 PS: This cover the basic concepts of cloud computing, system automation, high availability and web application hosting. If you know this to add Docker, CI/CD pipelines, Kubernetes, etc is a matter of just replacing parts of it. It's a solid ground work that can teach you a lot.