Ansible

Ansible’s rarely found on a dev shop’s list of technologies used; it’s traditionally thought of as a sysadmin or devops tool. But for us, it’s important enough to get its own page. Both our devs and sysadmins use it every day.

Ansible is an orchestration tool, like Chef or Puppet: It lets you run commands and make changes on multiple systems at once. What makes it different is that you don’t need to install it on those systems. All you need is SSH access. Because it’s mostly distro-independent, it works across the wide range of systems our clients use.

Parthenon became an early adopter of Ansible because it’s our philosophy that our dev and test machines should be as similar to production systems as we can get without actually testing in production. The virtual machines our guys are using to develop our clients’ code are deployed and created with the same configuration and set of software packages that the production systems are, which lets us test our software the way it’s actually going to behave when it goes live. Ansible is the core of that process; it guarantees that when we roll something out, the same set of steps are used that we went through during testing and client user-acceptance. Some shops need days of configuration and setup, or exotic container technologies, for multi-tier system rollout; with Ansible, our releases rarely take more than a few minutes.