add elasticache memory monitor

Author: kmackowick

aws/elasticache/README.md

@@ -40,6 +40,7 @@ No modules.
 | [datadog_monitor.hit_rate](https://registry.terraform.io/providers/datadog/datadog/latest/docs/resources/monitor) | resource |
 | [datadog_monitor.hit_rate_anomaly](https://registry.terraform.io/providers/datadog/datadog/latest/docs/resources/monitor) | resource |
 | [datadog_monitor.max_connections](https://registry.terraform.io/providers/datadog/datadog/latest/docs/resources/monitor) | resource |
+| [datadog_monitor.memory_utilization](https://registry.terraform.io/providers/datadog/datadog/latest/docs/resources/monitor) | resource |
 | [datadog_monitor.swap_usage](https://registry.terraform.io/providers/datadog/datadog/latest/docs/resources/monitor) | resource |
 
 ## Inputs
@@ -97,6 +98,12 @@ No modules.
 | <a name="input_max_connections_threshold_critical"></a> [max\_connections\_threshold\_critical](#input\_max\_connections\_threshold\_critical) | Critical threshold (connections) | `number` | `64000` | no |
 | <a name="input_max_connections_threshold_warning"></a> [max\_connections\_threshold\_warning](#input\_max\_connections\_threshold\_warning) | Warning threshold (connections) | `number` | `60000` | no |
 | <a name="input_max_connections_use_message"></a> [max\_connections\_use\_message](#input\_max\_connections\_use\_message) | Whether to use the query alert base message for max connections monitor | `bool` | `false` | no |
+| <a name="input_memory_utilization_enabled"></a> [memory\_utilization\_enabled](#input\_memory\_utilization\_enabled) | Enable memory utilization monitor | `bool` | `false` | no |
+| <a name="input_memory_utilization_evaluation_window"></a> [memory\_utilization\_evaluation\_window](#input\_memory\_utilization\_evaluation\_window) | Evaluation window for monitor (`last_?m` (1, 5, 10, 15, or 30), `last_?h` (1, 2, or 4), or `last_1d`) | `string` | `"last_1h"` | no |
+| <a name="input_memory_utilization_no_data_window"></a> [memory\_utilization\_no\_data\_window](#input\_memory\_utilization\_no\_data\_window) | No data threshold (in minutes, 0 to disable) | `number` | `15` | no |
+| <a name="input_memory_utilization_threshold_critical"></a> [memory\_utilization\_threshold\_critical](#input\_memory\_utilization\_threshold\_critical) | Critical threshold (percentage) | `number` | `80` | no |
+| <a name="input_memory_utilization_threshold_warning"></a> [memory\_utilization\_threshold\_warning](#input\_memory\_utilization\_threshold\_warning) | Warning threshold (percentage) | `number` | `70` | no |
+| <a name="input_memory_utilization_use_message"></a> [memory\_utilization\_use\_message](#input\_memory\_utilization\_use\_message) | Whether to use the query alert base message for memory utilization monitor | `bool` | `false` | no |
 | <a name="input_monitor_exclude_tags"></a> [monitor\_exclude\_tags](#input\_monitor\_exclude\_tags) | Tags to be excluded in the monitoring query. Specify in key:value format | `list(string)` | `[]` | no |
 | <a name="input_monitor_include_tags"></a> [monitor\_include\_tags](#input\_monitor\_include\_tags) | Tags to be included in the monitoring query. Specify in key:value format | `list(string)` | `[]` | no |
 | <a name="input_new_group_delay"></a> [new\_group\_delay](#input\_new\_group\_delay) | Delay in seconds before generating alerts for a new resource | `number` | `300` | no |
@@ -109,7 +116,7 @@ No modules.
 | <a name="input_notify_prod_override"></a> [notify\_prod\_override](#input\_notify\_prod\_override) | List of notifications for 12x5 prod alerts in critical threshold (uses `notify_default` otherwise) | `list(string)` | `[]` | no |
 | <a name="input_notify_recovery_override"></a> [notify\_recovery\_override](#input\_notify\_recovery\_override) | List of notifications for alert recovery (uses `notify_default` otherwise) | `list(string)` | `[]` | no |
 | <a name="input_notify_warn_override"></a> [notify\_warn\_override](#input\_notify\_warn\_override) | List of notifications for alerts in warning threshold (uses `notify_default` otherwise) | `list(string)` | `[]` | no |
-| <a name="input_renotify_interval"></a> [renotify\_interval](#input\_renotify\_interval) | Interval in minutes to re-send notifications about an alert | `number` | `0` | no |
+| <a name="input_renotify_interval"></a> [renotify\_interval](#input\_renotify\_interval) | Interval in minutes to re-send notifications about an alert | `number` | `60` | no |
 | <a name="input_runbook_link"></a> [runbook\_link](#input\_runbook\_link) | Runbook link to include in message | `string` | `null` | no |
 | <a name="input_service"></a> [service](#input\_service) | Service associated with the monitored resource (leave blank to omit tag) | `string` | `null` | no |
 | <a name="input_swap_usage_enabled"></a> [swap\_usage\_enabled](#input\_swap\_usage\_enabled) | Enable swap usage monitor | `bool` | `false` | no |

aws/elasticache/main.tf

@@ -212,3 +212,32 @@ END
     warning  = var.swap_usage_threshold_warning
   }
 }
+
+resource "datadog_monitor" "memory_utilization" {
+  count = var.memory_utilization_enabled ? 1 : 0
+
+  name         = join("", [local.title_prefix, "Elasticache Memory Utilization - {{replication_group.name}} - {{value}}%", local.title_suffix])
+  include_tags = false
+  message      = var.memory_utilization_use_message ? local.query_alert_base_message : ""
+  tags         = concat(local.common_tags, var.base_tags, var.additional_tags)
+  type         = "query alert"
+
+  evaluation_delay    = var.evaluation_delay
+  new_group_delay     = var.new_group_delay
+  notify_no_data      = var.notify_no_data
+  no_data_timeframe   = var.memory_utilization_no_data_window
+  renotify_interval   = var.renotify_interval
+  require_full_window = true
+  timeout_h           = var.timeout_h
+
+  query = <<END
+    avg(${var.memory_utilization_evaluation_window}):
+      avg:aws.elasticache.database_memory_usage_percentage${local.query_filter} by {replication_group,region,aws_account,env,datadog_managed}
+    >= ${var.memory_utilization_threshold_critical}
+END
+
+  monitor_thresholds {
+    critical = var.memory_utilization_threshold_critical
+    warning  = var.memory_utilization_threshold_warning
+  }
+}

aws/elasticache/variables.tf

@@ -321,3 +321,42 @@ variable "swap_usage_use_message" {
   type        = bool
   default     = false
 }
+
+########################################
+# Memory Utilization
+########################################
+variable "memory_utilization_enabled" {
+  default     = false
+  description = "Enable memory utilization monitor"
+  type        = bool
+}
+
+variable "memory_utilization_evaluation_window" {
+  default     = "last_1h"
+  description = "Evaluation window for monitor (`last_?m` (1, 5, 10, 15, or 30), `last_?h` (1, 2, or 4), or `last_1d`)"
+  type        = string
+}
+
+variable "memory_utilization_no_data_window" {
+  default     = 15
+  description = "No data threshold (in minutes, 0 to disable)"
+  type        = number
+}
+
+variable "memory_utilization_threshold_critical" {
+  default     = 80
+  description = "Critical threshold (percentage)"
+  type        = number
+}
+
+variable "memory_utilization_threshold_warning" {
+  default     = 70
+  description = "Warning threshold (percentage)"
+  type        = number
+}
+
+variable "memory_utilization_use_message" {
+  description = "Whether to use the query alert base message for memory utilization monitor"
+  type        = bool
+  default     = false
+}