All right, getting redash on kubernetes is like making a fish walk.
For Joule, we put most of our deployments on Kubernetes for the easy management and efficiency. But it does cause a bit of trouble with persistent deployments, especially redash. And this is how we solved it
It seems using a satefull prostgress DB on GKE is futile, we dicided to use GCP CloudSQL to sync with the postgress on every start up, so you can stop and delete the whole deployment and start it back up with all your information still on.
We are also very very lucky to have someone create most of the setup here. Shout out to Donald Rauscher.
Terraform, a HashiCorp software that helps you auto provision anything on GCP, AWS, and many many more.
Helm, Helm helps you manage Kubernetes applications — Helm Charts helps you define, install, and upgrade even the most complex Kubernetes application.
We will use Terraform to deploy our infrastructure, but before we do that we need to create a GCP service account and download it as a json.
After you downloaded the JSON, put it in a safe directory that you will remember, then set the environmental variable to GOOGLE_APPLICATION_CREDENTIALS
expose GOOGLE_APPLICATION_CREDENTIALS = /path/to/json Set the context
export PROJECT_ID=$(gcloud config get-value project -q)Run terraform
terraform apply -var project=${PROJECT_ID} # Set the base image
FROM redash/redash:4.0.0-rc.1
# Dockerfile author / maintainer
ENV REDASH_ADDITIONAL_QUERY_RUNNERS "redash.query_runner.cass"