Terraform Aws Alert Lambda Module

Lambda for Subscribing to an SNS Topic for CloudWatch Alarm from Metrics and sending it to Operations as the expected JSON object. See monitoring-json-object.md File for and Example...

New to Terraform Modules at BYU?

This module deploys a Lambda behind an ALB that receives Alarm Data from Metrics that exists in CloudWatch. That Alarm Data is then reformatted into the expected format that Operations is expecting to receive.

Usage

module "alert_lambda" {
  source = "github.com/byu-oit/terraform-aws-alert-lambda?ref=v1.1.1"
}

Requirements

• Terraform version 1.3.0 or greater
• AWS Provider 3.75.2 or greater
• BYU-ACS version 3.5.0 or greater

See examples/simple/simple-example.tf for a working example.

module "simple_test" {
  source                           = "../../"
  app_name                         = "alert-lambda-simple-test"
  monitoring_host                  = "https://in.monitoringdev.byu.edu/"
  kb                               = "KB000000"
  lambda_role_permissions_boundary = module.acs.role_permissions_boundary.arn
  vpc_id                           = module.acs.vpc.id
  subnet_ids                       = module.acs.private_subnet_ids
  metric_alarm_configs = [{
    alarm_name          = "simple-test-cpu-utilization-alarm"
    statistic           = "Average"
    metric_name         = "CPUUtilization"
    comparison_operator = "GreaterThanThreshold"
    threshold           = 85
    period              = 60
    evaluation_periods  = 1
    alarm_description   = "Alarm fires on cpu utilization"
    namespace           = "AWS/ECS"
    dimensions = {
      ServiceName = "tyk-identity-api-dev"
      ClusterName = "tyk-identity-api-dev"
    }
  }]
}

Inputs

Name	Type	Description	Default
app_name	string	The application name to include in the name of resources created.
kb	string	The KB article identifier for operations to use to resolve the alert. For example, KB0000000
monitoring_host	string	The host to ship the Monitoring JSON Object to.	https://in.monitoring.byu.edu
monitoring_path	string	The monitoring api path.	generic/process-event
lambda_role_permissions_boundary	string	The ARN of the role permissions boundary to attach to the Lambda role.	""
log_retention_in_days	number	The number of days to retain logs for the sns-to-teams Lambda.	7
memory_size	number	The amount of memory for the function.	128
timeout	number	The number of seconds the function is allowed to run.	30
tags	map(string)	A map of AWS Tags to attach to each resource created.	{}
metric_alarm_configs	list(map)	Array of Alarm objects	see example in variables.tf
vpc_id	string	Use a VPC for the lambda ingester functions. Pass in a vpc to enable.	""
security_group_ids	list(string)	A list of security group ids for the VPC configuration regarding the ingester lambda functions. Only required if VPC is enabled.	[]
period	number	(Optional) The period in seconds over which the specified statistic is applied.

"metric_alarm_configs" Object Types supported from variables.tf:53-67

variable "metric_alarm_configs" {
  type = list(object({
    alarm_name          = string
    statistic           = string
    metric_name         = string
    namespace           = string
    comparison_operator = string
    threshold           = number
    period              = number
    evaluation_periods  = number
    alarm_description   = string
    dimensions          = map(string)
  }))
  description = "Array of Alarm objects"
}

Outputs

Name	Type	Description

Future Modifications...

If you need to modify the standard SNS Topic for something special. See Terraform aws_sns_topic

There are several configurable Alarm Metric Alarm inputs that you can work with. See Terraform aws_cloudwatch_metric_alarm