A microservice that returns a country's geographical location information

I developed Get Country Data because I wanted to reduce the amount of costly calls my apps were making to the Google Geocoding/Places API and I wanted to know additional information like currency names, currency acronyms, currency symbols (HTML entities) and international telephone calling codes

The data came out of multiple other projects I was working on and is not read from any external source or a local database but rather it's kept in-memory

The Relational Database API allows access to MySQL over HTTP and ensures the price is low and predictable

This enables me to prevent direct access to the MySQL server and the query result table is returned in an associative array and is optionally serialized and stored in memcached

The API connects to the MySQL server using a user account that is granted all privileges on all databases and tables. It determines what user account to use by parsing the connection string read from the virtual host environment variables

The Relational Database Client is shipped with my PHP projects via Composer and JSON is the data interchange format

My build process works by concatenating all JS and pushing it through Babel and then through UglifyJS

Percy is invoked by the CI/CD pipeline and performs a pixel-diff

The JS and external stylesheets are cached by the browser and will be re-cached after a production deployment (the frontend web server sets the Cache-Control header)

The Google Cloud Storage API and client allows my stateless containerised apps to have persistent storage

It makes sense to ship this functionality as a microservice API because that way I can manage this logic in a central place for all my apps

I run all my app server infrastructure on cost effective Linode virtual machines and I do all the systems administration myself

The frontend and backend web servers run Apache HTTPd with mod_php and each run a local instance of memcached

All of my app backends are integrated into Papertrail, that I used for exception and general logging, and Papertrail invokes a Slack webhook and these exceptions are readable on my mobile phone

I host and maintain all my DNS zones myself using the BIND9 name server. Shell access is protected by private key and 2-factor authentication

I use Semaphore CI with all my apps and the pipeline is described in YAML and travels with the git repository and deploys into my static App Server Infrastructure

The deployment strategy is rolling and the webroot directory is a symlink which is updated after the source code is tarball'ed and transferred via SCP

I'm able to do plenty production deployments in a day without negatively effecting the users

For my frontend projects, Semaphore CI invokes Percy and all notifications are pushed into Slack

Puppet manages my app servers configuration and is directly applied on each server manually

There is no master and all the Puppet modules and manifests are in Git

Puppet manages everything from the user accounts and SSH server configuration to the Apache virtual hosts and PHP configuration

My App Server Infrastructure is in code and can be provisioned and managed with Terraform

If the infrastructure is newly created, Terraform will copy up and execute a shell script that installs Puppet and clones the Puppet Masterless Git repository and applies the configuration

This keeps my server infrastructure, configuration consistent and managed in code

I built a comprehensive Knative installation shell script that deploys Knative into managed Kubernetes from DigitalOcean

I can go from source code/container to a fully functional URL, with legitimate TLS, in seconds