Open kevinschaper opened 6 months ago
Should we set up a new repo to play around with this?
I wonder if there would be any security considerations to just putting the config into this project - that might be handy for updating infrastructure via GitHub actions
--edit to add, I'm also ok with a new repo or starting the process as a branch for refactoring monarch-stack-v3
I have created a new project for testing this called crispy-octo-fishstick-560367 in the TISlab GCP org. The name is a throw-away GitHub suggested when I made a repo for this exploration. I have it private for now but I'll share it with you if I make any interesting progress. I just wanted to mention the name in case you noticed the new project and had any concerns.
Let's set up a medium difficulty config for GKE (Google Kubernetes Engine) that runs the semsimian-server docker container that has phenio mounted and available to it.
It's probably worth starting here with a stateless hello world app
Then moving on to making a PersistentVolume & claiming it, and how we might be able to create volumes ahead of time and pre-load them (maybe useful)
Choosing semsimian-server is somewhat arbitrary, but we do want to sort out making files available to a container.
The next level will be multiple containers talking to one another, likely API+Solr, or API+semsimian-server.
I'm making this a group assignment so that we can work on it in it together.