You will use Docker + Docker Compose to install a support ticketing application (osTicket.com), a forum application (Flarum.org), and the LEMP stack

You will have fifteen days to complete this project. ✻ THE BORING PART ✻ You will install... osTicket (a support ticketing application) [login to view URL] Flarum (a forum application) [login to view URL] and, Nginx, PHP, and MariaDB (LEMP stack). ✻ THE INTERESTING PART ✻ You will provide me with a solution that will automatically create a new VPS (virtual private server) and change the DNS as a simple yet effective means of failover. For example, you might create a script that would create a new ticket every 15 minutes using the osTicket API.... [login to view URL] or the Flarum API.... [login to view URL] If osTicket or Flarum were to fail to respond by sending out a confirmation email within, say, 30 minutes then, for example, an Ansible playbook might be triggered to create a new *VPS*... Vultr [login to view URL] Hetzner [login to view URL] and change the *DNS*... Hetzner [login to view URL] Vultr [login to view URL] I don't know if Terraform would be necessary because Hetzner and Vultr provide VPSes with CentOS, Debian, Ubuntu, etc. ✻ OTHER DETAILS ✻ You will create a screencast as documentation so that I can replicate what you have done. You will have root access to a server running on a VPS at Hetzner and Vultr. I will provide documentation that you can read about this entire project. I'm not an engineer. However, I do realize that, "A little knowledge is a dangerous thing." I know enough to be dangerous. I need you to actually make this idea work safely and practically. ✻ HOW DID I COME UP WITH THIS SOLUTION? ✻ Here's what happened. I wearied of reading about Galera clusters, load balancing, Keepalived + HAProxy, split brain, high availability (HA), CloudFlare Pro, Docker Swarm, Kubernetes, and other solutions suitable for medium and larger enterprises. Yeah. I know. For Linux sys admins their transformation into DevOps gurus might be exciting. But, see, here's the thing: I'm not a DevOps guru. I aspire to know as little as possible about Linux system administration. I want to work with a sensible (practical) Linux sys admin (not a self-proclaimed DevOps guru) who will avoid providing me with unnecessary bells and whistles which I don't want and don't need. I want a practical solution that meets my needs. The inspiration for my idea came for this post I read on Reddit... [login to view URL] which included the following suggestion... ********** You, my dude or dudette, need immutable infrastructure in your life. Consider using hashicorp packer to make an immutable image of your server (usually just whatever is needed at runtime). Maybe use a managed distributed filesystem like EFS (Amazon) or equivalent so you can always just kill your server, spin up another one up from your immutable image, and mount the filesystem. Boom, you’re back in business.... ********** I took the, "You, my dude or dudette, need immutable infrastructure in your life", jettisoned (abandoned) Hashicorp's Packer and a managed distributed filesystem in favor of... IaaS [login to view URL] such as [login to view URL], [login to view URL], [login to view URL], etc. and I kept, "...so you can always just kill your server, spin up another one up from your immutable image, and mount the filesystem. Boom, you’re back in business." but instead of me "killing my server" I decided to automate that part so that I would have "cattle not pets." What is the definition of “cattle not pets”? [login to view URL]