VED-992: Metrics for mesh backlog (#1098)

* add custom metrics for mesh backlog objects

Author: motola

infrastructure/instance/environments/preprod/int-blue/variables.tfvars

@@ -3,5 +3,8 @@ immunisation_account_id           = "084828561157"
 dspp_core_account_id              = "603871901111"
 pds_environment                   = "int"
 error_alarm_notifications_enabled = true
+
+# mesh no invocation period metric set to 3 days (in seconds) for preprod environment i.e 3 * 24 * 60 * 60
+mesh_no_invocation_period_seconds = 259200
 create_mesh_processor             = true
 has_sub_environment_scope         = false

infrastructure/instance/environments/preprod/int-green/variables.tfvars

@@ -3,5 +3,8 @@ immunisation_account_id           = "084828561157"
 dspp_core_account_id              = "603871901111"
 pds_environment                   = "int"
 error_alarm_notifications_enabled = true
+
+# mesh no invocation period metric set to 3 days (in seconds) for preprod environment i.e 3 * 24 * 60 * 60
+mesh_no_invocation_period_seconds = 259200
 create_mesh_processor             = true
 has_sub_environment_scope         = false

infrastructure/instance/environments/prod/blue/variables.tfvars

@@ -3,6 +3,9 @@ immunisation_account_id           = "664418956997"
 dspp_core_account_id              = "232116723729"
 pds_environment                   = "prod"
 error_alarm_notifications_enabled = true
+
+# mesh no invocation period metric set to 1 day (in seconds) for prod environment i.e 1 * 24 * 60 * 60
+mesh_no_invocation_period_seconds = 86400
 create_mesh_processor             = true
 has_sub_environment_scope         = false
 dspp_kms_key_alias                = "nhsd-dspp-core-prod-extended-attributes-gdp-key"

infrastructure/instance/environments/prod/green/variables.tfvars

@@ -3,6 +3,9 @@ immunisation_account_id           = "664418956997"
 dspp_core_account_id              = "232116723729"
 pds_environment                   = "prod"
 error_alarm_notifications_enabled = true
+
+# mesh no invocation period metric set to 1 day (in seconds) for prod environment i.e 1 * 24 * 60 * 60
+mesh_no_invocation_period_seconds = 86400
 create_mesh_processor             = true
 has_sub_environment_scope         = false
 dspp_kms_key_alias                = "nhsd-dspp-core-prod-extended-attributes-gdp-key"

infrastructure/instance/mesh_processor.tf

@@ -288,3 +288,25 @@ resource "aws_cloudwatch_metric_alarm" "mesh_processor_error_alarm" {
   alarm_actions       = [data.aws_sns_topic.imms_system_alert_errors.arn]
   treat_missing_data  = "notBreaching"
 }
+
+resource "aws_cloudwatch_metric_alarm" "mesh_processor_no_lambda_invocation_alarm" {
+  count = var.create_mesh_processor && var.error_alarm_notifications_enabled ? 1 : 0
+
+  alarm_name        = "imms-${local.resource_scope}-mesh-processor-no-lambda-invocation"
+  alarm_description = "Triggers when the MESH Processor Lambda has no invocations for the configured time window."
+
+  metric_name = "Invocations"
+  namespace   = "AWS/Lambda"
+  statistic   = "Sum"
+  period      = var.mesh_no_invocation_period_seconds
+
+  evaluation_periods  = 1
+  comparison_operator = "LessThanThreshold"
+  threshold           = 1
+  treat_missing_data  = "breaching"
+  dimensions = {
+    FunctionName = aws_lambda_function.mesh_file_converter_lambda[0].function_name
+  }
+
+  alarm_actions = [data.aws_sns_topic.imms_system_alert_errors.arn]
+}

infrastructure/instance/variables.tf

@@ -40,6 +40,12 @@ variable "pds_environment" {
   default = "int"
 }
 
+variable "mesh_no_invocation_period_seconds" {
+  description = "The maximum duration the MESH Processor Lambda can go without being invoked before the no-invocation alarm is triggered."
+  type        = number
+  default     = 300
+}
+
 # Remember to switch off in PR envs after testing
 variable "error_alarm_notifications_enabled" {
   default     = true