In this post I will attempt to chronicle the steps my associates and I took to complete the configuration of static.grinnell.edu, and to eventually create this blog following Juan Treminio’s lead. Small portions of Juan’s blog post are reproduced here, with permission, so that you can follow along in his work.
Juan Treminio’s blog post does a nice job of covering the steps necessary to engage Watchtower, GitHub, and an automated build configuration in Docker Hub. The entire process can be used to push your initial Hugo project to production, watch for changes in your GitHub project repo, compile the changes, build a new Docker image, and automatically push it to production.
One of the really cool things I like about the workflow documented in Juan Treminio’s blog post is the ability to setup auto-build in Docker Hub. Unfortunately, that comes at a cost. Docker Hub’s ‘free’ account option will support only one parallel auto-build, so if you have more than one project you’d like to auto-build at Docker Hub you’ll need to pay for an account.
Have you ever wondered how a blog is born? The story behind this blog begins with my interest in stepping back from the CMS world, primarily Drupal, to discover the joys of static site generation. The journey begins in earnest at the 2016 DLF Forum: Milwaukee on the eve of the United States’ 2016 national election, when all the buzz that wasn’t political, was about building static web sites, and Jekyll.
Most of the servers I deploy to and manage here at Grinnell College are now “Dockerized”, and all of those use Traefik to manage traffic, of course. Before a web app or server can be opened for access to the world here, it has to pass a vulnerability scan, and I’m not privy to the specifics of that scan.
Had a conversation about lots of topics this afternoon and bumped into bleve, along the way, including a discussion about adding search capability to a Hugo site. bleeeve is based on the Go programming language, as is Hugo, so I hope implementation is super simple.