[eks_argo] Add event triggering support (#56)

* [eks_argo] Add event triggering support * Add some comments * refactor a bit - create a new file argo_events.tf * fix lints * push argo-events stuff into a module - with view to migrate it to metaflow-tools later
outerbounds · Apr 19, 2023 · 8974f22 · 8974f22
1 parent 5b27e96
commit 8974f22
Show file tree

Hide file tree

Showing 9 changed files with 281 additions and 6 deletions.
diff --git a/examples/eks_argo/argo_events.tf b/examples/eks_argo/argo_events.tf
@@ -0,0 +1,5 @@
+module "argo_events" {
+  depends_on     = [helm_release.argo]
+  source         = "./argo_events"
+  jobs_namespace = "default"
+}
diff --git a/examples/eks_argo/argo_events/argo-events-helper-chart/.helmignore b/examples/eks_argo/argo_events/argo-events-helper-chart/.helmignore
@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/
diff --git a/examples/eks_argo/argo_events/argo-events-helper-chart/Chart.yaml b/examples/eks_argo/argo_events/argo-events-helper-chart/Chart.yaml
@@ -0,0 +1,6 @@
+apiVersion: v2
+name: argo-events-helper-chart
+description: Helper chart that contains EventBus and EventSource definitions.
+type: application
+version: 0.1.0
+appVersion: "0.1.0"
diff --git a/examples/eks_argo/argo_events/argo-events-helper-chart/templates/eventbus.yaml b/examples/eks_argo/argo_events/argo-events-helper-chart/templates/eventbus.yaml
@@ -0,0 +1,16 @@
+apiVersion: argoproj.io/v1alpha1
+kind: EventBus
+metadata:
+  name: default
+  namespace: {{ .Values.jobsNamespace }}
+spec:
+  jetstream:
+    version: 2.9.15
+    containerTemplate:
+      resources:
+        limits:
+          cpu: 100m
+          memory: 128Mi
+        requests:
+          cpu: 100m
+          memory: 128Mi
diff --git a/examples/eks_argo/argo_events/argo-events-helper-chart/templates/eventsource.yaml b/examples/eks_argo/argo_events/argo-events-helper-chart/templates/eventsource.yaml
@@ -0,0 +1,24 @@
+apiVersion: argoproj.io/v1alpha1
+kind: EventSource
+metadata:
+  name: argo-events-webhook
+  namespace: {{ .Values.jobsNamespace }}
+spec:
+  template:
+    container:
+      resources:
+        requests:
+          cpu: 50m
+          memory: 50Mi
+        limits:
+          cpu: 50m
+          memory: 50Mi
+  service:
+    ports:
+      - port: 12000
+        targetPort: 12000
+  webhook:
+    metaflow-event:
+      port: "12000"
+      endpoint: /metaflow-event
+      method: POST
diff --git a/examples/eks_argo/argo_events/argo-events-helper-chart/values.yaml b/examples/eks_argo/argo_events/argo-events-helper-chart/values.yaml
@@ -0,0 +1 @@
+jobsNamespace: default
diff --git a/examples/eks_argo/argo_events/main.tf b/examples/eks_argo/argo_events/main.tf
@@ -0,0 +1,191 @@
+locals {
+  argo_events_values = {
+    "configs" = {
+      "jetstream" = {
+        "versions" = [
+          {
+            "configReloaderImage"  = "natsio/nats-server-config-reloader:latest"
+            "metricsExporterImage" = "natsio/prometheus-nats-exporter:latest"
+            "natsImage"            = "nats:latest"
+            "startCommand"         = "/nats-server"
+            "version"              = "latest"
+          },
+          {
+            "configReloaderImage"  = "natsio/nats-server-config-reloader:latest"
+            "metricsExporterImage" = "natsio/prometheus-nats-exporter:latest"
+            "natsImage"            = "nats:2.9.15"
+            "startCommand"         = "/nats-server"
+            "version"              = "2.9.15"
+          },
+        ]
+      }
+    }
+    "controller" = {
+      "name" = "controller-manager"
+      "rbac" = {
+        "enabled"    = true
+        "namespaced" = false
+      }
+      "resources" = {
+        "limits" = {
+          "cpu"    = "200m"
+          "memory" = "192Mi"
+        }
+        "requests" = {
+          "cpu"    = "200m"
+          "memory" = "192Mi"
+        }
+      }
+      "serviceAccount" = {
+        "create" = true
+        "name"   = "argo-events-events-controller-sa"
+      }
+    }
+    "crds" = {
+      "keep" = true
+    }
+    "extraObjects" = [
+      {
+        "apiVersion" = "v1"
+        "kind"       = "ServiceAccount"
+        "metadata" = {
+          "name"      = "operate-workflow-sa"
+          "namespace" = var.jobs_namespace
+        }
+      },
+      {
+        "apiVersion" = "rbac.authorization.k8s.io/v1"
+        "kind"       = "Role"
+        "metadata" = {
+          "name"      = "operate-workflow-role"
+          "namespace" = var.jobs_namespace
+        }
+        "rules" = [
+          {
+            "apiGroups" = [
+              "argoproj.io",
+            ]
+            "resources" = [
+              "workflows",
+              "workflowtemplates",
+              "cronworkflows",
+              "clusterworkflowtemplates",
+            ]
+            "verbs" = [
+              "*",
+            ]
+          },
+        ]
+      },
+      {
+        "apiVersion" = "rbac.authorization.k8s.io/v1"
+        "kind"       = "RoleBinding"
+        "metadata" = {
+          "name"      = "operate-workflow-role-binding"
+          "namespace" = var.jobs_namespace
+        }
+        "roleRef" = {
+          "apiGroup" = "rbac.authorization.k8s.io"
+          "kind"     = "Role"
+          "name"     = "operate-workflow-role"
+        }
+        "subjects" = [
+          {
+            "kind" = "ServiceAccount"
+            "name" = "operate-workflow-sa"
+          },
+        ]
+      },
+      {
+        "apiVersion" = "rbac.authorization.k8s.io/v1"
+        "kind"       = "Role"
+        "metadata" = {
+          "name"      = "view-events-role"
+          "namespace" = var.jobs_namespace
+        }
+        "rules" = [
+          {
+            "apiGroups" = [
+              "argoproj.io",
+            ]
+            "resources" = [
+              "eventsources",
+              "eventbuses",
+              "sensors",
+            ]
+            "verbs" = [
+              "get",
+              "list",
+              "watch",
+            ]
+          },
+        ]
+      },
+      {
+        "apiVersion" = "rbac.authorization.k8s.io/v1"
+        "kind"       = "RoleBinding"
+        "metadata" = {
+          "name"      = "view-events-role-binding"
+          "namespace" = var.jobs_namespace
+        }
+        "roleRef" = {
+          "apiGroup" = "rbac.authorization.k8s.io"
+          "kind"     = "Role"
+          "name"     = "view-events-role"
+        }
+        "subjects" = [
+          {
+            "kind"      = "ServiceAccount"
+            "name"      = "argo-workflows"
+            "namespace" = "argo-workflows"
+          },
+        ]
+      },
+    ]
+  }
+}
+
+resource "kubernetes_namespace" "argo_events" {
+  metadata {
+    name = "argo-events"
+  }
+}
+
+resource "helm_release" "argo_events" {
+  name = "argo-events"
+
+  repository   = "https://argoproj.github.io/argo-helm"
+  chart        = "argo-events"
+  namespace    = kubernetes_namespace.argo_events.metadata[0].name
+  force_update = true
+
+  values = [
+    yamlencode(local.argo_events_values)
+  ]
+}
+
+
+resource "helm_release" "argo_events_helper_chart" {
+  # We define an EventBus and EventSource in this helper chart. This is one
+  # of the cleaner workarounds for the chicken-egg problem with CR and CRD definitions
+  # in "terraform plan". E.g. Terraform tries to validate the kind "EventBus" before it
+  # has been created in the cluster, causing the validation to fail.
+  #
+  # Mega-thread here: https://github.com/hashicorp/terraform-provider-kubernetes/issues/1367
+  name = "argo-events-helper-chart"
+
+  depends_on = [helm_release.argo_events]
+
+  chart        = "${path.module}/argo-events-helper-chart"
+  namespace    = kubernetes_namespace.argo_events.metadata[0].name
+  force_update = true
+
+  set {
+    name  = "jobsNamespace"
+    value = var.jobs_namespace
+  }
+}
+
+variable "jobs_namespace" {
+  type = string
+}
diff --git a/examples/eks_argo/metaflow.tf b/examples/eks_argo/metaflow.tf
@@ -65,3 +65,8 @@ module "metaflow-metadata-service" {
 
   standard_tags = local.tags
 }
+
+variable "with_public_ip" {
+  type    = bool
+  default = true
+}
diff --git a/examples/eks_argo/metaflow_config.tf b/examples/eks_argo/metaflow_config.tf
@@ -5,13 +5,17 @@ data "aws_api_gateway_api_key" "metadata_api_key" {
 resource "local_file" "foo" {
   content = jsonencode({
     "METAFLOW_SERVICE_AUTH_KEY"           = data.aws_api_gateway_api_key.metadata_api_key.value
-    "METAFLOW_DATASTORE_SYSROOT_S3"       = module.metaflow-datastore.METAFLOW_DATASTORE_SYSROOT_S3,
-    "METAFLOW_DATATOOLS_S3ROOT"           = module.metaflow-datastore.METAFLOW_DATATOOLS_S3ROOT,
-    "METAFLOW_SERVICE_URL"                = module.metaflow-metadata-service.METAFLOW_SERVICE_URL,
-    "METAFLOW_KUBERNETES_NAMESPACE"       = "default",
-    "METAFLOW_KUBERNETES_SERVICE_ACCOUNT" = "argo-workflow",
-    "METAFLOW_DEFAULT_DATASTORE"          = "s3",
+    "METAFLOW_DATASTORE_SYSROOT_S3"       = module.metaflow-datastore.METAFLOW_DATASTORE_SYSROOT_S3
+    "METAFLOW_DATATOOLS_S3ROOT"           = module.metaflow-datastore.METAFLOW_DATATOOLS_S3ROOT
+    "METAFLOW_SERVICE_URL"                = module.metaflow-metadata-service.METAFLOW_SERVICE_URL
+    "METAFLOW_KUBERNETES_NAMESPACE"       = "default"
+    "METAFLOW_KUBERNETES_SERVICE_ACCOUNT" = "argo-workflow"
+    "METAFLOW_DEFAULT_DATASTORE"          = "s3"
     "METAFLOW_DEFAULT_METADATA"           = "service"
+    "METAFLOW_ARGO_EVENTS_EVENT_BUS"      = "default"
+    "METAFLOW_ARGO_EVENTS_EVENT_SOURCE"   = "argo-events-webhook"
+    "METAFLOW_ARGO_EVENTS_EVENT"          = "metaflow-event"
+    "METAFLOW_ARGO_EVENTS_WEBHOOK_URL"    = "http://argo-events-webhook-eventsource-svc.default:12000/metaflow-event"
   })
   filename = "${path.module}/config.json"
 }