Member-only story
Deploying HashiCorp Waypoint as a shared service
HashiCorp’s Waypoint is an innovative solution to deploying applications quickly to an environment while reducing the complexity often associated with a release process. As an application developer, you can deploy an application through a single command, waypoint up .
There are no makefiles, no CI/CD, just a single command, and within seconds the application is live. This statement is not to say the previously mentioned resources are outdated or no longer warranted. Waypoint is an alternative that focuses on the workflow versus the underlying technologies. And the workflow we are after is deploying an application as quick and straightforward as possible.
A good use case for Waypoint is to enable application developers to deploy applications to a test environment. With this use case in mind, there are not many examples of Waypoint being deployed from a platform owner’s perspective (shared service).
This article will explain the architecture and overall solution for deploying HashiCorp Waypoint as a shared service into your AWS environment. A GitHub project has been created for your convenience and is intended to be a starting point for you. Feel free to visit https://github.com/karl-cardenas-coding/waypoint-deployment-demo to get started.
Prerequisites
- HashiCorp Packer
- HashiCorp Waypoint
- HashiCorp Terraform
- AWS Account
- A custom domain
Architecture
The architecture is primarily composed of EC2s that are in an auto-scaling group. The Waypoint server is in an auto-scaling group with a max and min of 1. Additional servers increase the complexity due to having to share a common state/context. The Waypoint runners are needed to build, deploy and release applications. Without the runners, the end-users will have to build, deploy, and release the applications on their local workstations.
The runners can easily be scaled up through the auto-scaling group. However, without a service discovery…
