From cf043de3102a0ffdf4a7597fd5e2d9eb26e34eb7 Mon Sep 17 00:00:00 2001
From: Harriet H-W <harriethw@gmail.com>
Date: Fri, 24 Oct 2025 17:16:28 +0100
Subject: [PATCH] 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.
---
 infrastructure/modules/app-insights/alerts.tf | 26 ++++++++++++++++++
 infrastructure/modules/app-insights/tfdocs.md | 25 +++++++++++++++++
 .../modules/app-insights/variables.tf         | 27 +++++++++++++++++++
 3 files changed, 78 insertions(+)
 create mode 100644 infrastructure/modules/app-insights/alerts.tf

diff --git a/infrastructure/modules/app-insights/alerts.tf b/infrastructure/modules/app-insights/alerts.tf
new file mode 100644
index 00000000..337b6d13
--- /dev/null
+++ b/infrastructure/modules/app-insights/alerts.tf
@@ -0,0 +1,26 @@
+resource "azurerm_monitor_metric_alert" "exceptions" {
+  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
+  }
+}
diff --git a/infrastructure/modules/app-insights/tfdocs.md b/infrastructure/modules/app-insights/tfdocs.md
index 85d743de..ad2959f1 100644
--- a/infrastructure/modules/app-insights/tfdocs.md
+++ b/infrastructure/modules/app-insights/tfdocs.md
@@ -38,6 +38,30 @@ Type: `string`
 
 The following input variables are optional (have default values):
 
+### <a name="input_action_group_id"></a> [action\_group\_id](#input\_action\_group\_id)
+
+Description: The ID of the Action Group to use for alerts.
+
+Type: `string`
+
+Default: `null`
+
+### <a name="input_alert_frequency"></a> [alert\_frequency](#input\_alert\_frequency)
+
+Description: The frequency an alert is checked e.g. PT1M, PT5M, PT15M, PT30M, PT1H, PT6H, PT12H.
+
+Type: `string`
+
+Default: `"PT5M"`
+
+### <a name="input_enable_alerting"></a> [enable\_alerting](#input\_enable\_alerting)
+
+Description: Whether monitoring and alerting is enabled for this module.
+
+Type: `bool`
+
+Default: `false`
+
 ### <a name="input_tags"></a> [tags](#input\_tags)
 
 Description: A mapping of tags to assign to the resource.
@@ -70,3 +94,4 @@ Description: n/a
 The following resources are used by this module:
 
 - [azurerm_application_insights.appins](https://registry.terraform.io/providers/hashicorp/azurerm/latest/docs/resources/application_insights) (resource)
+- [azurerm_monitor_metric_alert.exceptions](https://registry.terraform.io/providers/hashicorp/azurerm/latest/docs/resources/monitor_metric_alert) (resource)
diff --git a/infrastructure/modules/app-insights/variables.tf b/infrastructure/modules/app-insights/variables.tf
index a4e575cf..3d671f22 100644
--- a/infrastructure/modules/app-insights/variables.tf
+++ b/infrastructure/modules/app-insights/variables.tf
@@ -33,3 +33,30 @@ 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 = var.alert_frequency
+}