Install with Terraform

If you are using Terraform to manage your Kubernetes infrastructure, use the Terraform Helm provider by HashiCorp or opentofu/helm provider to install Kedify Agent. The example installs the Agent, KEDA and HTTP Scaler.

Configure the Helm release

Obtain the organization ID and Agent API key first. Replace placeholders through your existing secret-management workflow. Terraform can retain credentials in state; use an access-controlled backend.

provider "helm" {
  kubernetes = {
    config_path = "~/.kube/config"
  }
}

resource "helm_release" "kedify-agent" {
  name              = "kedify-agent"
  repository        = "https://kedify.github.io/charts"
  chart             = "kedify-agent"
  version           = "v0.7.1"
  namespace         = "keda"
  wait              = false
  atomic            = false
  create_namespace  = true
  dependency_update = true

  values = [
    // file("${path.module}/agent-common-values.yaml"),
    yamlencode({
      clusterName = "my-cluster"
      agent = {
        orgId = "$YOUR_ORG_ID"
        apiKey = "$YOUR_API_KEY"
      }
      keda = {
        enabled = true
      }
      keda-add-ons-http = {
        enabled = true
      }
    })
  ]
}

With wait=false and atomic=false, Terraform does not wait for workload readiness or automatically roll back a failed Helm deployment. Run the checks below after applying. To enable Helm waiting or rollback, first verify hook completion with your selected chart and provider versions.

Apply and verify

Run terraform plan with your selected Helm provider and chart versions before applying.

terraform init
terraform plan
terraform apply

With wait=false, a successful apply does not establish component readiness. Run the installation checks.