fix: the alarm monitoring for 'unhealthyhost' wasn't working properly (…

…#614) * use arn suffix for cloud alarm setup * use Maximum instead of SampleCount
cds-snc · Mar 11, 2024 · 4309971 · 4309971
1 parent 02613ee
commit 4309971
Show file tree

Hide file tree

Showing 4 changed files with 22 additions and 17 deletions.
diff --git a/aws/alarms/cloudwatch.tf b/aws/alarms/cloudwatch.tf
@@ -67,12 +67,12 @@ resource "aws_cloudwatch_metric_alarm" "UnHealthyHostCount-TargetGroup1" {
   metric_name         = "UnHealthyHostCount"
   namespace           = "AWS/ApplicationELB"
   period              = "60"           # Every minute
-  statistic           = "SampleCount"  # use the number of data points during the period
+  statistic           = "Maximum"      # the highest value observed during the specified period
   treat_missing_data  = "notBreaching" # don't alarm if there's no data
   alarm_actions       = [var.sns_topic_alert_critical_arn]
   dimensions = {
     LoadBalancer = var.lb_arn_suffix
-    TargetGroup  = var.lb_target_group_1_arn
+    TargetGroup  = var.lb_target_group_1_arn_suffix
   }
 }
 
@@ -85,12 +85,12 @@ resource "aws_cloudwatch_metric_alarm" "UnHealthyHostCount-TargetGroup2" {
   metric_name         = "UnHealthyHostCount"
   namespace           = "AWS/ApplicationELB"
   period              = "60"           # Every minute
-  statistic           = "SampleCount"  # use the number of data points during the period
+  statistic           = "Maximum"      # the highest value observed during the specified period
   treat_missing_data  = "notBreaching" # don't alarm if there's no data
   alarm_actions       = [var.sns_topic_alert_critical_arn]
   dimensions = {
     LoadBalancer = var.lb_arn_suffix
-    TargetGroup  = var.lb_target_group_2_arn
+    TargetGroup  = var.lb_target_group_2_arn_suffix
   }
 }
 

diff --git a/aws/alarms/inputs.tf b/aws/alarms/inputs.tf
@@ -83,13 +83,13 @@ variable "lb_arn_suffix" {
   type        = string
 }
 
-variable "lb_target_group_1_arn" {
-  description = "Load balancer target group 1 ARN"
+variable "lb_target_group_1_arn_suffix" {
+  description = "Load balancer target group 1 ARN suffix, used by response time alarms"
   type        = string
 }
 
-variable "lb_target_group_2_arn" {
-  description = "Load balancer target group 2 ARN"
+variable "lb_target_group_2_arn_suffix" {
+  description = "Load balancer target group 2 ARN suffix, used by response time alarms"
   type        = string
 }
 

diff --git a/aws/load_balancer/outputs.tf b/aws/load_balancer/outputs.tf
@@ -18,9 +18,14 @@ output "lb_target_group_1_arn" {
   value       = aws_lb_target_group.form_viewer_1.arn
 }
 
-output "lb_target_group_2_arn" {
-  description = "Load balancer target group 2 ARN"
-  value       = aws_lb_target_group.form_viewer_2.arn
+output "lb_target_group_1_arn_suffix" {
+  description = "Load balancer target group 1 ARN suffix for use with CloudWatch Alarm"
+  value       = aws_lb_target_group.form_viewer_1.arn_suffix
+}
+
+output "lb_target_group_2_arn_suffix" {
+  description = "Load balancer target group 2 ARN suffix for use with CloudWatch Alarm"
+  value       = aws_lb_target_group.form_viewer_2.arn_suffix
 }
 
 output "lb_target_group_1_name" {

diff --git a/env/cloud/alarms/terragrunt.hcl b/env/cloud/alarms/terragrunt.hcl
@@ -38,8 +38,8 @@ dependency "load_balancer" {
   mock_outputs = {
     lb_arn                = null
     lb_arn_suffix         = null
-    lb_target_group_1_arn = null
-    lb_target_group_2_arn = null
+    lb_target_group_1_arn_suffix = null
+    lb_target_group_2_arn_suffix = null
   }
 }
 
@@ -108,10 +108,10 @@ inputs = {
   kms_key_cloudwatch_arn         = dependency.kms.outputs.kms_key_cloudwatch_arn
   kms_key_cloudwatch_us_east_arn = dependency.kms.outputs.kms_key_cloudwatch_us_east_arn
 
-  lb_arn                = dependency.load_balancer.outputs.lb_arn
-  lb_arn_suffix         = dependency.load_balancer.outputs.lb_arn_suffix
-  lb_target_group_1_arn = dependency.load_balancer.outputs.lb_target_group_1_arn
-  lb_target_group_2_arn = dependency.load_balancer.outputs.lb_target_group_2_arn
+  lb_arn                        = dependency.load_balancer.outputs.lb_arn
+  lb_arn_suffix                 = dependency.load_balancer.outputs.lb_arn_suffix
+  lb_target_group_1_arn_suffix  = dependency.load_balancer.outputs.lb_target_group_1_arn_suffix
+  lb_target_group_2_arn_suffix  = dependency.load_balancer.outputs.lb_target_group_2_arn_suffix
 
   sqs_reliability_deadletter_queue_arn = dependency.sqs.outputs.sqs_reliability_deadletter_queue_arn
   sqs_audit_log_deadletter_queue_arn   = dependency.sqs.outputs.sqs_audit_log_deadletter_queue_arn