feat: add support for suppressing notifications (#3)

Users can specify MSK cluster states that should not trigger a notification.
- Updated pre-commit hooks to use newer versions:
  - antonbabenko/[email protected]
  - pre-commit/[email protected]

.gitignore

@@ -26,3 +26,6 @@ terraform.rc
 
 # Python virtual environment
 .venv
+
+# Lambda zip directory
+out/

.pre-commit-config.yaml

@@ -1,6 +1,6 @@
 repos:
   - repo: https://github.com/antonbabenko/pre-commit-terraform
-    rev: v1.83.2
+    rev: v1.83.5
     hooks:
       - id: terraform_fmt
       - id: terraform_validate
@@ -27,7 +27,7 @@ repos:
         - --args=--quiet
         - --args=--skip-check CKV_AWS_116,CKV_AWS_117,CKV_AWS_173,CKV_AWS_272 
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v4.5.0
     hooks:
       - id: check-merge-conflict
       - id: end-of-file-fixer

README.md

@@ -51,6 +51,8 @@ No modules.
 | Name | Description | Type | Default | Required |
 |------|-------------|------|---------|:--------:|
 | <a name="input_email"></a> [email](#input\_email) | List of e-mail addresses subscribing to the SNS topic. Default is empty list. | `list(string)` | `[]` | no |
+| <a name="input_ignore_states"></a> [ignore\_states](#input\_ignore\_states) | Suppress warnings for the listed MSK states. Default: ['MAINTENANCE'] | `list(string)` | <pre>[<br>  "MAINTENANCE"<br>]</pre> | no |
+| <a name="input_log_retion_period_in_days"></a> [log\_retion\_period\_in\_days](#input\_log\_retion\_period\_in\_days) | Number of days logs will be retained. Default is 365 days. | `number` | `365` | no |
 | <a name="input_schedule_expression"></a> [schedule\_expression](#input\_schedule\_expression) | The schedule expression for the CloudWatch event rule. Default is 'rate(15 minutes)'. | `string` | `"rate(15 minutes)"` | no |
 | <a name="input_tags"></a> [tags](#input\_tags) | A map of tags to add to all resources. Default is empty map. | `map(string)` | `{}` | no |
 

data.tf

@@ -8,5 +8,5 @@ data "aws_region" "current" {}
 data "archive_file" "status_checker_code" {
   type        = "zip"
   source_dir  = "${path.module}/functions/check-msk-status/"
-  output_path = "${path.module}/python/hello-python.zip"
+  output_path = "${path.module}/out/check-msk-status.zip"
 }

examples/01_default_configuration/README.md

@@ -27,4 +27,4 @@ No inputs.
 ## Outputs
 
 No outputs.
-<!-- END_TF_DOCS -->
+<!-- END_TF_DOCS -->

functions/check-msk-status/index.py

@@ -1,37 +1,47 @@
 import boto3
 import os
 
+
 def lambda_handler(event, context):
-    LAMBDASNSTOPIC = os.environ['SNS_TOPIC_ARN']
-    region = 'eu-central-1'
+    LAMBDASNSTOPIC = os.environ["SNS_TOPIC_ARN"]
+    SUPPRESS_STATES = os.environ["SUPPRESS_STATES"].split(",")
+    region = "eu-central-1"
     # Create an MSK client
-    client = boto3.client('kafka', region_name=region)
+    client = boto3.client("kafka", region_name=region)
     # Retrieve a list of clusters
     response = client.list_clusters()
     # Extract the cluster ARNs from the response
-    cluster_arns = response['ClusterInfoList']
+    cluster_arns = response["ClusterInfoList"]
+
+    valid_states = ["ACTIVE"] + SUPPRESS_STATES
+    print(
+        "Notifications suppressed for these MSK states: {}".format(
+            ", ".join(valid_states)
+        )
+    )
 
     for cluster in cluster_arns:
-        arn = cluster['ClusterArn']
+        arn = cluster["ClusterArn"]
         response = client.describe_cluster(ClusterArn=arn)
-        status = response['ClusterInfo']['State']
-        sns_client = boto3.client('sns')
-
-        if status != 'ACTIVE':
-          print("The MSK cluster: {} needs attention.".format(arn))
-          sns_client.publish(TopicArn=LAMBDASNSTOPIC,
-                           Message="MSK cluster: " + arn + " needs attention. The status is: " + status,
-                           Subject="MSK Health Warning!")
+        status = response["ClusterInfo"]["State"]
+        print("The cluster is in state {}.".format(status))
+        sns_client = boto3.client("sns")
+        if status not in valid_states:
+            print("The MSK cluster: {} needs attention.".format(arn))
+            sns_client.publish(
+                TopicArn=LAMBDASNSTOPIC,
+                Message="MSK cluster: "
+                + arn
+                + " needs attention. The status is: "
+                + status,
+                Subject="MSK Health Warning!",
+            )
         else:
-          print(
-            "The MSK cluster: {} is in a healthy state, and is reachable and available for use.".format(
-                arn))
+            print(
+                "The MSK cluster: {} is in a healthy state, and is reachable and available for use.".format(
+                    arn
+                )
+            )
 
     # Return the status
-    return {
-        'statusCode': 200,
-        'body': 'OK'
-    }
-
-if __name__ == '__main__':
-    lambda_handler(None, None)
+    return {"statusCode": 200, "body": "OK"}

main.tf

@@ -20,8 +20,7 @@ resource "aws_sns_topic_subscription" "msk_health_sns_topic_email_target" {
 
 # IAM role
 resource "aws_iam_role" "msk_health_lambda_role" {
-  name = "msk-health-lambda-role-${random_id.id.hex}"
-
+  name               = "msk-health-lambda-role-${random_id.id.hex}"
   assume_role_policy = <<EOF
 {
   "Version": "2012-10-17",
@@ -55,8 +54,7 @@ resource "aws_iam_policy" "msk_health_lambda_role_policy" {
   name        = "msk-health-lambda-role-policy-${random_id.id.hex}"
   path        = "/"
   description = "IAM policy msk health solution lambda"
-
-  policy = <<EOF
+  policy      = <<EOF
 {
     "Version": "2012-10-17",
     "Statement": [
@@ -87,12 +85,11 @@ resource "aws_iam_policy" "msk_health_lambda_role_policy" {
     ]
 }
 EOF
-  tags   = var.tags
+  tags        = var.tags
 }
 
-
 resource "aws_lambda_function" "msk_health_lambda" {
-  filename                       = "${path.module}/python/hello-python.zip"
+  filename                       = data.archive_file.status_checker_code.output_path
   function_name                  = "msk_status_monitor-${random_id.id.hex}"
   description                    = "MSK Status Monitor"
   role                           = aws_iam_role.msk_health_lambda_role.arn
@@ -102,10 +99,13 @@ resource "aws_lambda_function" "msk_health_lambda" {
   memory_size                    = 128
   timeout                        = 60
   tags                           = var.tags
-
+  tracing_config {
+    mode = "Active"
+  }
   environment {
     variables = {
-      SNS_TOPIC_ARN = aws_sns_topic.msk_health_sns_topic.arn
+      SNS_TOPIC_ARN   = aws_sns_topic.msk_health_sns_topic.arn
+      SUPPRESS_STATES = join(",", var.ignore_states)
     }
   }
 
@@ -137,6 +137,6 @@ resource "aws_lambda_permission" "allow_cw_call_lambda" {
 # Log group for the Lambda function
 resource "aws_cloudwatch_log_group" "msk_health_lambda_log_groups" {
   name              = "/aws/lambda/msk_status_monitor-${random_id.id.hex}"
-  retention_in_days = 30
+  retention_in_days = var.log_retion_period_in_days
   tags              = var.tags
 }

variables.tf

@@ -4,11 +4,32 @@ variable "email" {
   default     = []
 }
 
+variable "ignore_states" {
+  description = "Suppress warnings for the listed MSK states. Default: ['MAINTENANCE']"
+  type        = list(string)
+  default = [
+    "MAINTENANCE"
+  ]
+}
+
+variable "log_retion_period_in_days" {
+  type        = number
+  default     = 365
+  description = "Number of days logs will be retained. Default is 365 days."
+
+  validation {
+    condition = contains([1, 3, 5, 7, 14, 30, 60, 90, 120, 150, 180, 365,
+    400, 545, 731, 1096, 1827, 2192, 2557, 2992, 3288, 3653], var.log_retion_period_in_days)
+    error_message = "log_retion_period_in_days must be one of the allowed values: 1, 3, 5, 7, 14, 30, 60, 90, 120, 150, 180, 365, 400, 545, 731, 1096, 1827, 2192, 2557, 2922, 3288, 3653"
+  }
+}
+
 variable "schedule_expression" {
   description = "The schedule expression for the CloudWatch event rule. Default is 'rate(15 minutes)'."
   type        = string
   default     = "rate(15 minutes)"
 }
+
 variable "tags" {
   description = "A map of tags to add to all resources. Default is empty map."
   type        = map(string)