Merge pull request #12 from rhythmictech/ENG-4329

ensure rds used_storage monitor excludes aurora instances

Author: sdickenson

.pre-commit-config.yaml

@@ -10,7 +10,7 @@ repos:
       - id: terraform_validate
         exclude: common
         args:
-        - --hook-config=--retry-once-with-cleanup=true
+          - --hook-config=--retry-once-with-cleanup=true
       - id: terraform_tflint
         alias: terraform_tflint_nocreds
         name: terraform_tflint_nocreds

.tflint.hcl

@@ -1,5 +1,5 @@
 config {
-  module     = true
+  call_module_type     = "all"
 }
 
 plugin "aws" {

aws/rds/README.md

@@ -44,10 +44,10 @@ No modules.
 | <a name="input_alert_critical_priority"></a> [alert\_critical\_priority](#input\_alert\_critical\_priority) | Priority for alerts within critical threshold (P1-P5, uses monitor defaults if not specified) | `string` | `null` | no |
 | <a name="input_alert_message"></a> [alert\_message](#input\_alert\_message) | Message to prepend to alert notifications | `string` | `"Alert"` | no |
 | <a name="input_alert_nodata_priority"></a> [alert\_nodata\_priority](#input\_alert\_nodata\_priority) | Priority for alerts within warning threshold (P1-P5, uses monitor defaults if not specified) | `string` | `null` | no |
-| <a name="input_base_tags"></a> [base\_tags](#input\_base\_tags) | Base tags (key:value format) to add to this type of check (combined with `local.tags` and `var.additional_tags`, generally you should not change this) | `list(string)` | <pre>[<br>  "resource:rds"<br>]</pre> | no |
+| <a name="input_base_tags"></a> [base\_tags](#input\_base\_tags) | Base tags (key:value format) to add to this type of check (combined with `local.tags` and `var.additional_tags`, generally you should not change this) | `list(string)` | <pre>[<br/>  "resource:rds"<br/>]</pre> | no |
 | <a name="input_connection_count_anomaly_deviations"></a> [connection\_count\_anomaly\_deviations](#input\_connection\_count\_anomaly\_deviations) | Standard deviations | `number` | `3` | no |
 | <a name="input_connection_count_anomaly_enabled"></a> [connection\_count\_anomaly\_enabled](#input\_connection\_count\_anomaly\_enabled) | Enable CPU utilization anomaly monitor | `bool` | `true` | no |
-| <a name="input_connection_count_anomaly_evaluation_window"></a> [connection\_count\_anomaly\_evaluation\_window](#input\_connection\_count\_anomaly\_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_connection_count_anomaly_evaluation_window"></a> [connection\_count\_anomaly\_evaluation\_window](#input\_connection\_count\_anomaly\_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_4h"` | no |
 | <a name="input_connection_count_anomaly_no_data_window"></a> [connection\_count\_anomaly\_no\_data\_window](#input\_connection\_count\_anomaly\_no\_data\_window) | No data threshold (in minutes, 0 to disable) | `number` | `10` | no |
 | <a name="input_connection_count_anomaly_recovery_window"></a> [connection\_count\_anomaly\_recovery\_window](#input\_connection\_count\_anomaly\_recovery\_window) | Recovery window for anomaly monitor (`last_?m` (1, 5, 10, 15, or 30), `last_?h` (1, 2, or 4), or `last_1d`] | `string` | `"last_15m"` | no |
 | <a name="input_connection_count_anomaly_rollup"></a> [connection\_count\_anomaly\_rollup](#input\_connection\_count\_anomaly\_rollup) | Rollup interval (must be sized based on evaluation window/span and seasonaility) | `number` | `60` | no |
@@ -89,7 +89,7 @@ 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_team"></a> [team](#input\_team) | Team supporting the monitored resource (leave blank to omit tag) | `string` | `null` | no |

aws/rds/main.tf

@@ -1,11 +1,17 @@
 locals {
   # these must be defined but do not need to be overridden
-  monitor_alert_default_priority  = null
-  monitor_warn_default_priority   = null
+  # tflint-ignore: terraform_unused_declarations
+  monitor_alert_default_priority = null
+  # tflint-ignore: terraform_unused_declarations
+  monitor_warn_default_priority = null
+  # tflint-ignore: terraform_unused_declarations
   monitor_nodata_default_priority = null
 
   title_prefix = var.title_prefix == null ? "" : "[${var.title_prefix}]"
   title_suffix = var.title_suffix == null ? "" : " (${var.title_suffix})"
+
+  # Add RDS-specific query filter to exclude
+  rds_query_filter = "{!engine:aurora*,${trimprefix(local.query_filter, "{")}"
 }
 
 resource "datadog_monitor" "connection_count_anomaly" {
@@ -129,8 +135,8 @@ resource "datadog_monitor" "used_storage" {
   query = <<END
     max(${var.used_storage_evaluation_window}):(
       100 - ((
-        default(avg:aws.rds.free_storage_space${local.query_filter} by {dbinstanceidentifier,region,aws_account,env,datadog_managed}, 0) /
-        default(avg:aws.rds.total_storage_space${local.query_filter} by {dbinstanceidentifier,region,aws_account,env,datadog_managed}, 1)
+        default(avg:aws.rds.free_storage_space${local.rds_query_filter} by {dbinstanceidentifier,region,aws_account,env,datadog_managed}, 0) /
+        default(avg:aws.rds.total_storage_space${local.rds_query_filter} by {dbinstanceidentifier,region,aws_account,env,datadog_managed}, 1)
       ) * 100)
     ) >= ${var.used_storage_threshold_critical}
 END