From 9c4f5b4cfd652ec2b4a0e82454807b396f4a621c Mon Sep 17 00:00:00 2001
From: Alexandre Gaillet <alexandre.gaillet@fr.clara.net>
Date: Sat, 26 May 2018 02:41:37 +0200
Subject: [PATCH] MON-32 - Memory usage monitor updated

---
 cloud/aws/elasticache/common/inputs.tf        |  2 +-
 cloud/aws/elasticache/memcached/locals.tf     | 46 +++++++++----------
 .../memcached/monitors-memcached.tf           | 10 ++--
 cloud/aws/elasticache/redis/inputs.tf         |  2 +-
 cloud/aws/elasticache/redis/monitors-redis.tf | 14 +++---
 5 files changed, 39 insertions(+), 35 deletions(-)

diff --git a/cloud/aws/elasticache/common/inputs.tf b/cloud/aws/elasticache/common/inputs.tf
index ba43b3d..d174c2f 100644
--- a/cloud/aws/elasticache/common/inputs.tf
+++ b/cloud/aws/elasticache/common/inputs.tf
@@ -61,7 +61,7 @@ variable "max_connection_message" {
 variable "max_connection_time_aggregator" {
   description = "Monitor aggregator for Elasticache max connection [available values: min, max or avg]"
   type        = "string"
-  default     = "min"
+  default     = "max"
 }
 
 variable "max_connection_timeframe" {
diff --git a/cloud/aws/elasticache/memcached/locals.tf b/cloud/aws/elasticache/memcached/locals.tf
index 310de51..b044b56 100644
--- a/cloud/aws/elasticache/memcached/locals.tf
+++ b/cloud/aws/elasticache/memcached/locals.tf
@@ -1,27 +1,27 @@
 locals {
   memory = {
-    cache.t2.micro    = "595926712"
-    cache.t2.small    = "1664299827"
-    cache.t2.medium   = "3457448673"
-    cache.m3.medium   = "2985002270"
-    cache.m3.large    = "6496138035"
-    cache.m3.xlarge   = "14280766259"
-    cache.m3.2xlarge  = "29957396889"
-    cache.m4.large    = "6893422510"
-    cache.m4.xlarge   = "15333033246"
-    cache.m4.2xlarge  = "31890132172"
-    cache.m4.4xlarge  = "65262028062"
-    cache.m4.10xlarge = "166043435663"
-    cache.r3.large    = "14495514624"
-    cache.r3.xlarge   = "30494267801"
-    cache.r3.2xlarge  = "62491774156"
-    cache.r3.4xlarge  = "126701535232"
-    cache.r3.8xlarge  = "254476812288"
-    cache.r4.large    = "13207024435"
-    cache.r4.xlarge   = "26897232691"
-    cache.r4.2xlarge  = "54191749857"
-    cache.r4.4xlarge  = "108855946117"
-    cache.r4.8xlarge  = "218248763146"
-    cache.r4.16xlarge = "437012922368"
+    cache.t2.micro    = 595926712
+    cache.t2.small    = 1664299827
+    cache.t2.medium   = 3457448673
+    cache.m3.medium   = 2985002270
+    cache.m3.large    = 6496138035
+    cache.m3.xlarge   = 14280766259
+    cache.m3.2xlarge  = 29957396889
+    cache.m4.large    = 6893422510
+    cache.m4.xlarge   = 15333033246
+    cache.m4.2xlarge  = 31890132172
+    cache.m4.4xlarge  = 65262028062
+    cache.m4.10xlarge = 166043435663
+    cache.r3.large    = 14495514624
+    cache.r3.xlarge   = 30494267801
+    cache.r3.2xlarge  = 62491774156
+    cache.r3.4xlarge  = 126701535232
+    cache.r3.8xlarge  = 254476812288
+    cache.r4.large    = 13207024435
+    cache.r4.xlarge   = 26897232691
+    cache.r4.2xlarge  = 54191749857
+    cache.r4.4xlarge  = 108855946117
+    cache.r4.8xlarge  = 218248763146
+    cache.r4.16xlarge = 437012922368
   }
 }
diff --git a/cloud/aws/elasticache/memcached/monitors-memcached.tf b/cloud/aws/elasticache/memcached/monitors-memcached.tf
index 9754cb1..027a43d 100644
--- a/cloud/aws/elasticache/memcached/monitors-memcached.tf
+++ b/cloud/aws/elasticache/memcached/monitors-memcached.tf
@@ -117,13 +117,15 @@ resource "datadog_monitor" "memcached_free_memory" {
   name    = "[${var.environment}] Elasticache memcached free memory {{#is_alert}}{{{comparator}}} {{threshold}}% ({{value}}%){{/is_alert}}{{#is_warning}}{{{comparator}}} {{warn_threshold}}% ({{value}}%){{/is_warning}}"
   message = "${coalesce(var.free_memory_message, var.message)}"
 
+  count = "${length(keys(local.memory))}"
+
   type = "metric alert"
 
   query = <<EOF
     ${var.free_memory_time_aggregator}(${var.free_memory_timeframe}): (
-      avg:aws.elasticache.freeable_memory{${data.template_file.filter.rendered}} by {region,cacheclusterid,cachenodeid} /
-      ${local.memory[var.elasticache_size]} * 100
-    ) < ${var.free_memory_threshold_critical}
+      avg:aws.elasticache.freeable_memory{dd_monitoring:enabled,dd_aws_red:enabled,env:${var.environment},cache_node_type:${element(keys(local.memory), count.index)}} by {region,cacheclusterid,cachenodeid} /
+      ${element(values(local.memory), count.index)}
+    ) * 100 < ${var.free_memory_threshold_critical}
   EOF
 
   thresholds {
@@ -131,7 +133,7 @@ resource "datadog_monitor" "memcached_free_memory" {
     critical = "${var.free_memory_threshold_critical}"
   }
 
-  notify_no_data      = true
+  notify_no_data      = false
   evaluation_delay    = "${var.delay}"
   renotify_interval   = 0
   notify_audit        = false
diff --git a/cloud/aws/elasticache/redis/inputs.tf b/cloud/aws/elasticache/redis/inputs.tf
index 890f502..3c0380c 100644
--- a/cloud/aws/elasticache/redis/inputs.tf
+++ b/cloud/aws/elasticache/redis/inputs.tf
@@ -78,7 +78,7 @@ variable "cpu_high_message" {
 variable "cpu_high_time_aggregator" {
   description = "Monitor aggregator for Elasticache redis cpu high [available values: min, max or avg]"
   type        = "string"
-  default     = "min"
+  default     = "avg"
 }
 
 variable "cpu_high_timeframe" {
diff --git a/cloud/aws/elasticache/redis/monitors-redis.tf b/cloud/aws/elasticache/redis/monitors-redis.tf
index dddb799..f51e1ef 100644
--- a/cloud/aws/elasticache/redis/monitors-redis.tf
+++ b/cloud/aws/elasticache/redis/monitors-redis.tf
@@ -26,7 +26,7 @@ resource "datadog_monitor" "redis_cache_hits" {
       avg:aws.elasticache.cache_hits{${data.template_file.filter.rendered}} by {region,cacheclusterid}.as_count() /
       (avg:aws.elasticache.cache_hits{${data.template_file.filter.rendered}} by {region,cacheclusterid}.as_count() +
         avg:aws.elasticache.cache_misses{${data.template_file.filter.rendered}} by {region,cacheclusterid}.as_count())
-    ) < ${var.cache_hits_threshold_critical}
+    ) * 100 < ${var.cache_hits_threshold_critical}
   EOF
 
   thresholds {
@@ -145,8 +145,8 @@ resource "datadog_monitor" "redis_commands" {
 
   query = <<EOF
     sum(${var.commands_timeframe}): (
-      avg:aws.elasticache.get_type_cmds{${data.template_file.filter.rendered}} by {region,cacheclusterid,cachenodeid}.as_count() +
-      avg:aws.elasticache.set_type_cmds{${data.template_file.filter.rendered}} by {region,cacheclusterid,cachenodeid}.as_count()
+      avg:aws.elasticache.get_type_cmds{${data.template_file.filter.rendered}} by {region,cacheclusterid}.as_count() +
+      avg:aws.elasticache.set_type_cmds{${data.template_file.filter.rendered}} by {region,cacheclusterid}.as_count()
     ) <= 0
   EOF
 
@@ -169,12 +169,14 @@ resource "datadog_monitor" "redis_free_memory" {
   name    = "[${var.environment}] Elasticache redis free memory {{#is_alert}}{{{comparator}}} {{threshold}}% ({{value}}%){{/is_alert}}{{#is_warning}}{{{comparator}}} {{warn_threshold}}% ({{value}}%){{/is_warning}}"
   message = "${coalesce(var.free_memory_message, var.message)}"
 
+  count = "${length(keys(local.memory))}"
+
   type = "metric alert"
 
   query = <<EOF
     ${var.free_memory_time_aggregator}(${var.free_memory_timeframe}): (
-      avg:aws.elasticache.freeable_memory{${data.template_file.filter.rendered}} by {region,cacheclusterid,cachenodeid} /
-      ( ${local.memory[var.elasticache_size]} / ${var.nodes} )
+      avg:aws.elasticache.freeable_memory{dd_monitoring:enabled,dd_aws_red:enabled,env:${var.environment},cache_node_type:${element(keys(local.memory), count.index)}} by {region,cacheclusterid,cachenodeid} /
+      ( ${element(values(local.memory), count.index)} / ${var.nodes} )
     ) * 100 < ${var.free_memory_threshold_critical}
   EOF
 
@@ -183,7 +185,7 @@ resource "datadog_monitor" "redis_free_memory" {
     critical = "${var.free_memory_threshold_critical}"
   }
 
-  notify_no_data      = true
+  notify_no_data      = false
   evaluation_delay    = "${var.delay}"
   renotify_interval   = 0
   notify_audit        = false