Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

Hi All , working on setting up Flyte on GKE as part of a side project.
Does anyone have a document on how to deploy the helm chart on GKE cluster using helm_release resource block of terraform?
I'm trying to make sure all the changes are tracked via terraform for me to easily pull apart the infra after use.

```resource "helm_release" "flyte" {
  name    = "flyte"
  version = local.flyte_version

  namespace = "flyte"

  repository = "<https://helm.flyte.org>"
  chart      = "flyte-core"

  values = [
    "${file(.../values-gcp.yaml")}"
  ]

  set {
    name  = "userSettings.dbHost"
    value = var.x
  }
  set {
    name  = "userSettings.dbPassword"
    value = var.x
  }

  set {
    name  = "userSettings.hostName"
    value = var.x
  }

}```

Thanks <@U04664Z7H37> for the assist :v: , I made the changes and ran the actions pipeline , ran into error while connecting to GKE for deploying the helm chart , sorting that one out too. Letting you know soon

Hi Fabio , the terraform part threw a TLS error while applying so I used basic helm commands and deployed flyte binary without any values as overrides and the pod got scheduled and is waiting for DB , currently working on how to fix the helm provider part to set the DB creds and details and not expose it via values.yaml

I have a helm values file with place holders for the credentials and in the terraform helm release I use additional `set`s to overwrite these place holders in the values file with tf vars.

Yes this helps too , I can reference the sensitive vars using set block of helm_release and non sensitive ones through values.yaml file

ohh wait , I misread it , can you give a sample example of the approach , I think I have seen it somewhere

I was able to get the helm_release work by adding the cluster_ca_certificate field in helm provider which I was missing and create_namespace=true in the helm_release resource block.

Thanks Fabio , I'm now checking the crashloopbackoff issue which is showing not able to connect to DB whereas its open and was allowed the Cluster External IP