Merge pull request #5 from oracle-quickstart/feature/monitoring

add monitoring stack
oracle-quickstart · Mar 29, 2023 · dbcaaf9 · dbcaaf9
2 parents cf0e1f9 + ad8675a
commit dbcaaf9
Show file tree

Hide file tree

Showing 20 changed files with 1,529 additions and 22 deletions.
diff --git a/add_on_dependency_matrix.tf b/add_on_dependency_matrix.tf
@@ -0,0 +1,12 @@
+## Copyright © 2023, Oracle and/or its affiliates. 
+## All rights reserved. The Universal Permissive License (UPL), Version 1.0 as shown at http://oss.oracle.com/licenses/upl
+
+# defines trigger to enable specific components based on selection
+locals {
+  enable_flink                    = var.enable_flink
+  enable_cert_manager             = local.enable_flink || var.enable_cert_manager
+  enable_cluster_autoscaler       = var.np1_enable_autoscaler || var.np2_enable_autoscaler || var.np3_enable_autoscaler
+  enable_monitoring_stack         = var.enable_monitoring_stack
+  enable_metrics_server           = local.enable_cluster_autoscaler || var.enable_metrics_server || local.enable_monitoring_stack
+  enable_grafana_flink_dashboards = local.enable_monitoring_stack && local.enable_flink
+}
diff --git a/helm_cert_manager.tf → add_on_helm_cert_manager.tf b/helm_cert_manager.tf → add_on_helm_cert_manager.tf
@@ -4,9 +4,9 @@
 ## https://github.com/jetstack/cert-manager/blob/master/README.md
 ## https://artifacthub.io/packages/helm/cert-manager/cert-manager
 
-locals {
-  enable_cert_manager = var.enable_flink ? true : var.enable_cert_manager
-}
+# locals {
+#   enable_cert_manager = var.enable_flink ? true : var.enable_cert_manager
+# }
 
 resource "helm_release" "cert_manager" {
   count            = local.enable_cert_manager ? 1 : 0

diff --git a/helm_flink.tf → add_on_helm_flink.tf b/helm_flink.tf → add_on_helm_flink.tf
diff --git a/k8s_autoscaler.tf → add_on_k8s_autoscaler.tf b/k8s_autoscaler.tf → add_on_k8s_autoscaler.tf
diff --git a/add_on_monitoring_stack.tf b/add_on_monitoring_stack.tf
@@ -0,0 +1,103 @@
+# Copyright (c) 2021, 2023, Oracle and/or its affiliates. All rights reserved.
+# Licensed under the Universal Permissive License v 1.0 as shown at http://oss.oracle.com/licenses/upl. 
+
+locals {
+  deployment_name     = "kps"
+  vars                = { "region" = var.region, "tenancy_ocid" = var.tenancy_ocid }
+  scrape_configs      = flatten([for i in fileset("${path.module}/templates", "*.scrapeConfigs.yaml") : file("${path.module}/templates/${i}")])
+  grafana_datasources = flatten([for i in fileset("${path.module}/templates", "grafana.*.datasource.yaml") : yamldecode(templatefile("${path.module}/templates/${i}", local.vars))])
+  grafana_dashboards  = flatten([for i in fileset("${path.module}/templates", "grafana.*.dashboard.json") : { "name" = i, "label" = split(".", i)[1] }])
+  grafana_plugins     = file("${path.module}/templates/grafana.plugins.yaml")
+}
+
+output dash {
+  value = local.grafana_dashboards
+}
+
+resource "random_password" "grafana_password" {
+  count            = local.enable_monitoring_stack ? 1 : 0
+  length           = 20
+  special          = true
+  override_special = "#$%&@!_+=./;:][{}]"
+}
+
+output "grafana_password" {
+  value     = local.enable_monitoring_stack ? random_password.grafana_password[0].result : ""
+  sensitive = true
+}
+
+resource "helm_release" "kube_prometheus_stack" {
+  count            = local.enable_monitoring_stack ? 1 : 0
+  name             = local.deployment_name
+  repository       = "https://prometheus-community.github.io/helm-charts"
+  chart            = "kube-prometheus-stack"
+  namespace        = "monitoring"
+  version          = "45.8.0"
+  wait             = false
+  create_namespace = true
+
+  set {
+    name  = "prometheus.prometheusSpec.additionalScrapeConfigs"
+    value = join("\n", local.scrape_configs)
+  }
+
+  # set {
+  #   name  = "grafana.sidecar.datasources.defaultDatasourceEnabled"
+  #   value = false
+  # }
+
+  set {
+    name  = "grafana.adminPassword"
+    value = random_password.grafana_password[0].result
+  }
+
+  values = [
+    yamlencode({ "grafana" = {
+      "additionalDataSources" = local.grafana_datasources,
+      "plugins"               = yamldecode(local.grafana_plugins)
+    } })
+  ]
+
+  depends_on = [
+    oci_containerengine_node_pool.oci_oke_node_pool
+  ]
+}
+
+resource "kubernetes_config_map_v1" "grafana_dashboards" {
+  count = local.enable_monitoring_stack ? length(local.grafana_dashboards) : 0
+
+  metadata {
+    name      = "${local.deployment_name}-grafana-${local.grafana_dashboards[count.index].label}"
+    namespace = "monitoring"
+    labels = {
+      "grafana_dashboard" = "1"
+    }
+  }
+
+  data = {
+    "${local.grafana_dashboards[count.index].name}" = "${file("${path.module}/templates/${local.grafana_dashboards[count.index].name}")}"
+  }
+  depends_on = [
+    helm_release.kube_prometheus_stack
+  ]
+}
+
+resource "kubernetes_config_map_v1" "grafana_plugins" {
+  count = local.enable_monitoring_stack ? 1 : 0
+
+  metadata {
+    name      = "${local.deployment_name}-grafana-plugins"
+    namespace = "monitoring"
+    labels = {
+      "grafana_plugin" = "1"
+    }
+  }
+
+  data = {
+    "plugins" = local.grafana_plugins
+  }
+
+  depends_on = [
+    helm_release.kube_prometheus_stack
+  ]
+}
diff --git a/helm_metrics.tf → add_onn_helm_metrics.tf b/helm_metrics.tf → add_onn_helm_metrics.tf
@@ -1,9 +1,9 @@
 ## Copyright © 2022-2023, Oracle and/or its affiliates. 
 ## All rights reserved. The Universal Permissive License (UPL), Version 1.0 as shown at http://oss.oracle.com/licenses/upl
 
-locals {
-  enable_metrics_server = var.np1_enable_autoscaler || var.np2_enable_autoscaler || var.np3_enable_autoscaler ? true : var.enable_metrics_server
-}
+# locals {
+#   enable_metrics_server = var.np1_enable_autoscaler || var.np2_enable_autoscaler || var.np3_enable_autoscaler ? true : var.enable_metrics_server
+# }
 
 resource "helm_release" "metrics_server" {
   count      = local.enable_metrics_server ? 1 : 0

diff --git a/datasources.tf b/datasources.tf
@@ -1,4 +1,4 @@
-## Copyright © 2022, Oracle and/or its affiliates. 
+## Copyright © 2022-2023, Oracle and/or its affiliates. 
 ## All rights reserved. The Universal Permissive License (UPL), Version 1.0 as shown at http://oss.oracle.com/licenses/upl
 
 data "oci_containerengine_cluster_option" "cluster_options" {
@@ -12,8 +12,7 @@ data "oci_containerengine_node_pool_option" "oci_oke_node_pool_option" {
 # Gets home and current regions
 data "oci_identity_tenancy" "tenant_details" {
   tenancy_id = var.tenancy_ocid
-
-  provider = oci.current_region
+  provider   = oci.current_region
 }
 
 data "oci_identity_regions" "home_region" {
@@ -49,6 +48,7 @@ data "oci_limits_limit_definitions" "limit_def" {
   service_name   = "compute"
 }
 
+# buidl maps of valid shapes for each AD
 locals {
   availability_map = [for def in data.oci_limits_limit_definitions.limit_def.limit_definitions : def if contains(compact([var.np1_node_shape, var.np2_node_shape, var.np3_node_shape]), def.description)]
   limits_definitions = [
@@ -75,6 +75,7 @@ data "oci_core_shapes" "valid_shapes" {
   availability_domain = data.oci_identity_availability_domains.ADs.availability_domains[count.index].name
 }
 
+# Deploy ID to uniquely identify this cluster and associated resources.
 resource "random_string" "deploy_id" {
   length      = 4
   special     = false

diff --git a/images.tf b/images.tf
@@ -15,6 +15,7 @@ data "oci_core_image" "np3_image" {
   image_id = var.np3_image_id
 }
 
+# Identify if an OKE specific image is available for the Compute image selected
 locals {
   k8s_version = replace(local.kubernetes_version, "v", "")
   np1_oke_image = var.node_pool_count >= 1 ? [for option
@@ -30,12 +31,3 @@ locals {
   option if length(regexall("${data.oci_core_image.np3_image[0].display_name}-OKE-${local.k8s_version}", option.source_name)) > 0] : []
   np3_oke_image_id = length(local.np3_oke_image) > 0 ? local.np3_oke_image[0].image_id : var.np3_image_id
 }
-
-# output "images" {
-#   value = {
-#     k8s_version   = local.k8s_version
-#     np1_oke_image = local.np1_oke_image
-#     np2_oke_image = local.np2_oke_image
-#     np3_oke_image = local.np3_oke_image
-#   }
-# }
diff --git a/outputs.tf b/outputs.tf
@@ -61,7 +61,7 @@ output "access_command" {
 }
 
 output "flink_demo_job" {
-  value = "kubectl create -f https://raw.githubusercontent.com/apache/flink-kubernetes-operator/release-1.2/examples/basic.yaml"
+  value = "kubectl create -f https://raw.githubusercontent.com/apache/flink-kubernetes-operator/release-1.3/examples/basic.yaml"
 }
 
 output "flink_ui_port_forward" {

diff --git a/policies.tf b/policies.tf
@@ -18,7 +18,7 @@ locals {
 }
 
 resource "oci_identity_network_source" "node_pool_network_source" {
-  provider       = oci.home_region
+  provider = oci.home_region
   #Required
   compartment_id = var.tenancy_ocid
   description    = "NSG for ${local.nsg_name} autoscaler"

diff --git a/schema.yaml b/schema.yaml
@@ -161,6 +161,7 @@ variableGroups:
       - enable_flink
       - enable_cert_manager
       - enable_metrics_server
+      - enable_monitoring_stack
 
 variables:
 
@@ -1069,13 +1070,21 @@ variables:
     description: |
       Apache Flink will be installed using the Flink Operator.
 
+  enable_monitoring_stack:
+    type: boolean
+    default: true
+    title: Deploy Monitoring Stack
+    description: |
+      Deploys Prometheus, Grafana and related datasources, plugins and dashboards.
+
 outputGroups:
-  - title: Access Command
+  - title: Access
     outputs: 
       - access_command
       - flink_demo_job
       - flink_ui_port_forward
       - flink_ui_access
+      - grafana_password
 
 outputs:
   access_command:
@@ -1100,4 +1109,10 @@ outputs:
     type: link
     title: Flink UI
     displayText: "Access the Flink UI"
-    visible: true
+    visible: true
+
+  grafana_password:
+    type: copyableString
+    title: Grafana access
+    displayText: "Grafana password for 'admin' user"
+    visible: 
diff --git a/templates/grafana.default.dashboards.yaml b/templates/grafana.default.dashboards.yaml
@@ -0,0 +1,44 @@
+k8s-cluster:
+  gnetId: 7249
+  revision: 1
+  datasource: Prometheus
+k8s-cluster-metrics:
+  gnetId: 11663
+  revision: 1
+  datasource: Prometheus
+k8s-cluster-metrics-simple: 
+  gnetId: 6417
+  revision: 1
+  datasource: Prometheus
+k8s-pods-monitoring: 
+  gnetId: 13498
+  revision: 1
+  datasource: Prometheus
+k8s-memory: 
+  gnetId: 13421
+  revision: 1
+  datasource: Prometheus
+k8s-networking: 
+  gnetId: 12658
+  revision: 1
+  datasource: Prometheus
+k8s-cluster-autoscaler:
+  gnetId: 3831
+  revision: 1
+  datasource: Prometheus
+k8s-hpa:
+  gnetId: 10257
+  revision: 1
+  datasource: Prometheus
+k8s-pods:
+  gnetId: 6336
+  revision: 1
+  datasource: Prometheus
+oci-compute:
+  gnetId: 13596
+  revision: 1
+  datasource: Oracle Cloud Infrastructure Metrics
+oci-oke:
+  gnetId: 13594
+  revision: 1
+  datasource: Oracle Cloud Infrastructure Metrics