WIP

aks/application/data.tf

@@ -0,0 +1,19 @@
+data "azurerm_resource_group" "monitoring" {
+  count = var.azure_enable_container_monitoring ? 1 : 0
+
+  name = "${var.azure_resource_prefix}-${var.service_short}-mn-rg"
+}
+
+data "azurerm_monitor_action_group" "main" {
+  count = var.azure_enable_container_monitoring ? 1 : 0
+
+  name                = "${var.azure_resource_prefix}-${var.service_name}"
+  resource_group_name = data.azurerm_resource_group.monitoring[0].name
+}
+
+data "azurerm_kubernetes_cluster" "main" {
+  count = var.azure_enable_container_monitoring ? 1 : 0
+
+  name                = "${var.cluster_configuration_map.resource_prefix}-aks"
+  resource_group_name = var.cluster_configuration_map.resource_group_name
+}

aks/application/resources.tf

@@ -244,3 +244,37 @@ resource "kubernetes_secret" "ghcr_auth" {
     })
   }
 }
+
+resource "azurerm_monitor_metric_alert" "container_restarts" {
+  count = var.azure_enable_container_monitoring ? 1 : 0
+
+  name                = "${local.app_name}-restarts"
+  resource_group_name = data.azurerm_resource_group.monitoring[0].name
+  scopes              = [data.azurerm_kubernetes_cluster.main[0].id]
+  description         = "Action will be triggered when container restarts greate than 0"
+  window_size         = "PT30M"
+
+  criteria {
+    metric_namespace = "Insights.container/pods"
+    metric_name      = "restartingContainerCount"
+    aggregation      = "Maximum"
+    operator         = "GreaterThan"
+    threshold        = 0
+
+  dimension {
+      name     = "controllerName"
+      operator = "Include"
+      values   = ["${local.app_name}"]
+    }
+  }
+
+  action {
+    action_group_id = data.azurerm_monitor_action_group.main[0].id
+  }
+
+  lifecycle {
+    ignore_changes = [
+      tags
+    ]
+  }
+}

aks/application/variables.tf

@@ -102,3 +102,20 @@ variable "github_personal_access_token" {
   default     = null
   description = "Github Personal Access Token (PAT) of github_username"
 }
+
+variable "azure_resource_prefix" {
+  type        = string
+  default     = null
+  description = "Prefix of Azure resources for the service"
+}
+
+variable "service_short" {
+  type        = string
+  default     = null
+  description = "Short name of the service"
+}
+
+variable "azure_enable_container_monitoring" {
+  type    = bool
+  default = false
+}