Skip to content

Latest commit

 

History

History
73 lines (41 loc) · 3.7 KB

README.md

File metadata and controls

73 lines (41 loc) · 3.7 KB
Raw

example-hasura

Auth Example's Hasura instance, configs, migrations, and job to run migrations.

This particular example relies on SchemaHero to manage the "readmodel" database. However, if you'd like to use Hasura to manage this DB's migrations, simply uncomment the following lines:

https://github.com/CloudNativeEntrepreneur/example-hasura/blob/master/Dockerfile#L7 https://github.com/CloudNativeEntrepreneur/example-hasura/blob/master/apply-migrations.sh#L6

Environment Variabls

HASURA_GRAPHQL_ENDPOINT

Using this env when running Hasura console will override the value in config.yaml.

HASURA_GRAPHQL_ENDPOINT=http://some-endpoint hasura console

To get the Hasura url, run kubectl get ksvc in the namespace.

NAME                           URL                                                                LATESTCREATED                        LATESTREADY                          READY   REASON
example-hasura                https://example-hasura.jx-staging.cloudnativeentrepreneur.com                 example-hasura-00008                example-hasura-00008                True

Model services

To add a new model service as an available env variable for Hasura actions, edit charts/example-hasura/templates/ksvc.yaml

What the Chart creates

This chart provisions a PSQL cluster with two databases, hasura owned by the user hasura, and sourced owned by the user sourced, as well as a knative serving service that runs the hasura-graphql-engine.

Local Development with local-dev-cluster

For local development, the easiest way to use Hasura is to rely on the database in the local development cluster, while running Hasura through docker-compose - this was Hasura can send Actions to your local microservices, but still have the database created and maintained via Schemahero.

To do so:

  1. Make sure local-dev-cluster is running: https://github.com/CloudNativeEntrepreneur/local-dev-cluster

This will set up Postgres Operator and Schemahero.

  1. Deploy a Database and Schema to your local dev cluster: https://github.com/CloudNativeEntrepreneur/example-readmodel

In that repo, the command make onboard is configured to deploy the db to your local cluster.

  1. "Port Forward" the database so it's accessible locally

In a new terminal window, as you need to leave the process running, run:

kubectl port-forward example-readmodel-postgresql-0 5433:5432
  1. Start Hasura in this repo against port-forwarded db

Run make up to start Hasura via docker-compose configured to connect to the port-forwarded database. This command runs a script which looks up the required password for the db and provides it to the docker-compose.yaml file.

  1. When adding actions in HASURA, first add an ENV VAR to the docker-compose file with the URL, using host.docker.internal instead of localhost as the host. Using ENV Vars is not required, but makes switching to production versions simple later on.

It's worth noting that you could port-forward any psql db running in any kubernetes cluster. It does not have to be the local-dev-cluster, or the example-readmodel db, however, development being simple is important, and hence I've preconfigured this solution for development purposes.

FAQ

YOUR SECRET IS EXPOSED

Yes, I know.

It'll only work in the local development cluster - this is part of an example that contains several moving parts, so I just generated some random secrets where they were needed and preconfigured things accordingly so you can just run it locally and everything will work. Couldn't go without secrets as part of that example is an authentication server and it's JWT integration with Hasura.

Don't use these proconfigured values in production. I typically use ExternalSecrets in prod.