Add an alert on Exceptions for App Insights

We want to trigger an alert whenever there is an
exception for manage-breast-screening -
specifically for the Invite team's container app
jobs in the notifications app.

This should trigger whenever the Exception count
is above 0 - may want to tweak that but in theory
there should never be an Exception unless
something has completely failed.

Author: Harriethw

infrastructure/modules/app-insights/alerts.tf

@@ -0,0 +1,26 @@
+resource "azurerm_monitor_metric_alert" "res-0" {
+  count = var.enable_alerting ? 1 : 0
+
+  auto_mitigate            = true
+  description              = "Triggered by any Exception"
+  enabled                  = true
+  frequency                = var.alert_frequency
+  name                     = "Exceptions"
+  resource_group_name      = var.resource_group_name
+  scopes                   = [azurerm_application_insights.appins.id]
+  severity                 = 1
+  window_size              = local.alert_window_size
+
+  action {
+    action_group_id    = var.action_group_id
+  }
+
+  criteria {
+    aggregation            = "Count"
+    metric_name            = "exceptions/count"
+    metric_namespace       = "microsoft.insights/components"
+    operator               = "GreaterThan"
+    skip_metric_validation = false
+    threshold              = 0
+  }
+}

infrastructure/modules/app-insights/variables.tf

@@ -33,3 +33,36 @@ variable "resource_group_name" {
   type        = string
   description = "The name of the resource group in which the App Insights is created. Changing this forces a new resource to be created."
 }
+
+variable "enable_alerting" {
+  description = "Whether monitoring and alerting is enabled for this module."
+  type        = bool
+  default     = false
+}
+
+variable "alert_frequency" {
+  type     = string
+  nullable = true
+  default  = "PT5M"
+  validation {
+    condition     = contains(["PT1M", "PT5M", "PT15M", "PT30M", "PT1H"], var.alert_frequency)
+    error_message = "The alert_frequency must be one of: PT1M, PT5M, PT15M, PT30M, PT1H"
+  }
+  description = "The frequency an alert is checked e.g. PT1M, PT5M, PT15M, PT30M, PT1H, PT6H, PT12H."
+}
+
+variable "action_group_id" {
+  type        = string
+  description = "The ID of the Action Group to use for alerts."
+  default     = null
+}
+
+locals {
+  alert_window_size_map = {
+    PT5M  = "PT6M"
+    PT15M = "PT16M"
+    PT30M = "PT31M"
+    PT1H  = "PT61M"
+  }
+  alert_window_size = local.alert_window_size_map[var.alert_frequency]
+}