Development today relies on multiple teams, services, and environments all working in unison. A topic that always comes up, when setting up a new development environment: How do we secure important credentials, while not making it too complicated for the rest of the team?
The key when working with version control systems like Git, is to keep any type of credentials out of the versioning system. These can be API keys, database or email passwords.
Even if its a private repository, development environments might change. It can be a simple staging & live website setup you are maintaining.
The simplest way in PHP is to use .env files to store your credentials outside of the public accessible directory structure. So outside the public_html, but still within the reach of the executing environment to read it. Variables are accessible through $_ENV['yourVar'] or getenv("yourVar"), once included in your code.
To make it simple you can use the popular package vlucas/phpdotenv, which reads and imports the file automatically.
Don’t fool yourself, if an attacker finds a way into your system, these variables can be easily read. This is just hiding the file from public access and provides some convenience while developing or sharing code.
Some people propose to encrypt / decrypt environment variables using a secret key. But if an attacker can access your data, he can also find the secret key.
There are some nice packages that offer just that. You have to decide if those fit your ammo.
php artisan credentials:edit
The Apache2 environment variables are set in the /etc/apache2/envvars file. These variables are not the same as the environment variables of your Linux system; they are stored and manipulated in an internal Apache structure.
The /etc/apache2/envvars file holds variable definitions such as APACHE_LOG_DIR (the location of Apache log files), APACHE_PID_FILE (the Apache process ID), APACHE_RUN_USERS (the user that run Apache, by defaultwww-data), etc.
You can open and modify this file in a text editor of your choice. This is nice, but far from simple and requires a server restart. This is something which helps you when hardening security on a live deployed setup.
There are dynamic approaches, but you can do some research for that yourself :) Skipped that rabbit hole for now …
Handling secrets completely detached is another possibility. This is surely an overkill for most cases, but using an Infrastructure Secret Management concept might be worth looking into, if you are working on bigger scale projects that involve multiple development teams and setups. These services also often deal with secret rotation.
HashiCorp Vault – “Vault is a tool for securely accessing secrets. A secret is anything that you want to tightly control access to, such as API keys, passwords, certificates, and more. Vault provides a unified interface to any secret, while providing tight access control and recording a detailed audit log.”
You can deploy your own vault on your own infrastructure or test out a hosted version, which is free for Open Source projects. HashiCorp Vault
You will find a bunch of Hashicorp related packages that will help you to integrate a vault into your project workflow (scmrus/php-vault-env , poc-webapp-vault).
While this is nice, you will need to cache / store credentials somewhere, as you don’t want to query the vault on every single access.
The Hashicorp Vault is not the only Infrastructure Secret Management solution. There is a nice Github Gist that lists other solutions and a nice feature matrix.
Amazon also provides a solution called AWS Secrets Manager, which makes a lot of sense, when you build and deploy on AWS already :)
Gatsby is a free and open source framework based on React that helps developers build blazing fast websites and apps.
While researching some popular static site generation tools, GatsbyJS comes up often. I have played with NuxtJS and Hugo in the past, but what I REALLY like about GatsbyJS is the plugin / modular system. You can build your website with plain-old React and CSS styles, but make your development more efficient by adding node_modules.
Also being able to import any data source with ease, using GraphQL, is amazing. And when it comes to content management, you can easily hook a headless WordPress or Drupal setup into the mix and consume their REST APIs :)
I am not switching my own website to GatsbyJS anytime soon, but its another tool in my toolbox for future project consideration !
There are many tutorials on Youtube about getting started, maybe something to consider for the next freetime testdrive ;) Enjoy …
GatsbyJS @ Github
Robo 3T gives you the full power of the MongoDB shell, but also allows you to easily save, load, edit your queries, do copy/paste, view your results as JSON/trees/tables… and get a decent autocompletion!
It is cross-platform, free and open-source (GitHub repo here).