Add container app job alerts

Author: mrlockstar

infrastructure/modules/container-app-job/alerts.tf

@@ -0,0 +1,30 @@
+resource "azurerm_monitor_scheduled_query_rules_alert" "job_failure" {
+  count = var.enable_alerting ? 1 : 0
+
+  name                = "${azurerm_container_app_job.this.name}-containerjob-failure-alert"
+  location            = var.location
+  resource_group_name = var.resource_group_name
+
+  action {
+    action_group     = [ var.action_group_id ]
+  }
+  data_source_id      = var.log_analytics_workspace_id
+  description         = "Alert when ${azurerm_container_app_job.this.name} job fails"
+  enabled             = true
+  query          = <<-KQL
+    ContainerAppSystemLogs_CL
+    | where ExecutionName_s contains "${azurerm_container_app_job.this.name}"
+    | where Log_s !contains "JobCleanup: Pod"
+    | where Reason_s == "ProcessExited"
+    | extend ExitCode = toint(extract(@"exit code:\\s*(\\d+)", 1, Log_s))
+    | where ExitCode != 0
+    | summarize Failures = count() by ExecutionName_s
+  KQL
+  severity            = 1
+  frequency           = var.alert_frequency
+  time_window         = var.time_window
+  trigger {
+    threshold          = 0
+    operator           = "GreaterThan"
+  }
+}

infrastructure/modules/container-app-job/tfdocs.md

@@ -16,6 +16,12 @@ Description: Docker image and tag. Format: <registry>/<repository>:<tag>
 
 Type: `string`
 
+### <a name="input_log_analytics_workspace_audit_id"></a> [log\_analytics\_workspace\_audit\_id](#input\_log\_analytics\_workspace\_audit\_id)
+
+Description: Log analytics workspace audit ID
+
+Type: `string`
+
 ### <a name="input_name"></a> [name](#input\_name)
 
 Description: Name of the container app. Limited to 32 characters
@@ -48,6 +54,30 @@ Type: `string`
 
 Default: `null`
 
+### <a name="input_action_group_id"></a> [action\_group\_id](#input\_action\_group\_id)
+
+Description: ID of the action group to notify.
+
+Type: `string`
+
+Default: `null`
+
+### <a name="input_alert_frequency"></a> [alert\_frequency](#input\_alert\_frequency)
+
+Description: Frequency (in minutes) at which rule condition should be evaluated. Values must be between 5 and 1440 (inclusive). Default is 15
+
+Type: `number`
+
+Default: `15`
+
+### <a name="input_alert_window_size"></a> [alert\_window\_size](#input\_alert\_window\_size)
+
+Description: The period of time that is used to monitor alert activity e.g. PT1M, PT5M, PT15M, PT30M, PT1H, PT6H, PT12H. The interval between checks is adjusted accordingly.
+
+Type: `string`
+
+Default: `"PT5M"`
+
 ### <a name="input_app_key_vault_id"></a> [app\_key\_vault\_id](#input\_app\_key\_vault\_id)
 
 Description: ID of the key vault to store app secrets. Each secret is mapped to an environment variable. Required when fetch\_secrets\_from\_app\_key\_vault is true.
@@ -80,6 +110,14 @@ Type: `string`
 
 Default: `null`
 
+### <a name="input_enable_alerting"></a> [enable\_alerting](#input\_enable\_alerting)
+
+Description: Whether monitoring and alerting is enabled for the PostgreSQL Flexible Server.
+
+Type: `bool`
+
+Default: `false`
+
 ### <a name="input_environment_variables"></a> [environment\_variables](#input\_environment\_variables)
 
 Description: Environment variables to pass to the container app. Only non-secret variables. Secrets must be stored in key vault 'app\_key\_vault\_id'
@@ -154,6 +192,14 @@ Type: `map(string)`
 
 Default: `{}`
 
+### <a name="input_time_window"></a> [time\_window](#input\_time\_window)
+
+Description: Time window for which data needs to be fetched for query (must be greater than or equal to frequency). Values must be between 5 and 2880 (inclusive).  Default is 30
+
+Type: `number`
+
+Default: `30`
+
 ### <a name="input_user_assigned_identity_ids"></a> [user\_assigned\_identity\_ids](#input\_user\_assigned\_identity\_ids)
 
 Description: List of user assigned identity IDs to assign to the container app.
@@ -190,4 +236,5 @@ Version:
 The following resources are used by this module:
 
 - [azurerm_container_app_job.this](https://registry.terraform.io/providers/hashicorp/azurerm/latest/docs/resources/container_app_job) (resource)
+- [azurerm_monitor_scheduled_query_rules_alert.job_failure](https://registry.terraform.io/providers/hashicorp/azurerm/latest/docs/resources/monitor_scheduled_query_rules_alert) (resource)
 - [azurerm_key_vault_secrets.app](https://registry.terraform.io/providers/hashicorp/azurerm/latest/docs/data-sources/key_vault_secrets) (data source)

infrastructure/modules/container-app-job/variables.tf

@@ -125,6 +125,46 @@ variable "workload_profile_name" {
   nullable    = false
 }
 
+variable "enable_alerting" {
+  description = "Whether monitoring and alerting is enabled for the PostgreSQL Flexible Server."
+  type        = bool
+  default     = false
+}
+
+variable "action_group_id" {
+  type        = string
+  description = "ID of the action group to notify."
+  default     = null
+}
+
+variable "alert_window_size" {
+  type     = string
+  nullable = false
+  default  = "PT5M"
+  validation {
+    condition     = contains(["PT1M", "PT5M", "PT15M", "PT30M", "PT1H", "PT6H", "PT12H"], var.alert_window_size)
+    error_message = "The alert_window_size must be one of: PT1M, PT5M, PT15M, PT30M, PT1H, PT6H, PT12H"
+  }
+  description = "The period of time that is used to monitor alert activity e.g. PT1M, PT5M, PT15M, PT30M, PT1H, PT6H, PT12H. The interval between checks is adjusted accordingly."
+}
+
+variable "log_analytics_workspace_id" {
+  description = "Log analytics workspace ID"
+  type        = string
+}
+
+variable "alert_frequency" {
+  type        = number
+  description = "Frequency (in minutes) at which rule condition should be evaluated. Values must be between 5 and 1440 (inclusive). Default is 15"
+  default     = 15
+}
+
+variable "time_window" {
+  type        = number
+  description = "Time window for which data needs to be fetched for query (must be greater than or equal to frequency). Values must be between 5 and 2880 (inclusive).  Default is 30"
+  default     = 30
+}
+
 locals {
   memory = "${var.memory}Gi"
   cpu    = var.memory / 2