check_cloudwatch_metric

Icinga/Nagios check to test an AWS CloudFormation metric against thresholds.

Required Modules

• boto3

Installation

# Checkout source
git clone https://github.com/elias5000/check_cloudwatch_metric.git

# Install boto3 Python module
pip install boto3

# Copy check script
cp check_cloudwatch_metric.py /usr/lib/nagios/plugins/check_cloudwatch_metric.py

# Copy director config
cp check_cloudwatch_metric.conf /etc/icinga2/conf.d/check_cloudwatch_metric.conf

Authentication

Authentication is identical to awscli. Use either instance role EC2 or pod role on K8S with kube2iam (preferred) or ~/.aws/config profile. The check will use the default profile.

Commandline Usage

usage: check_cloudwatch_metric.py [-h] --name NAME --namespace NAMESPACE
                                  --warning WARNING --critical CRITICAL
                                  [--dimensions DIMENSIONS] [--last_state]
                                  [--minutes MINUTES] [--prefix PREFIX]
                                  [--region REGION] [--statistics STATISTICS]

optional arguments:
  -h, --help            show this help message and exit
  --dimensions DIMENSIONS
                        metric dimensions in form "Name1:Value1,Name2:Value2"
  --last_state          use last known value
  --minutes MINUTES     time window to aggregate for statistic
  --prefix PREFIX       metric namespace prefix (default: AWS)
  --profile PROFILE     AWS config profile
  --region REGION       AWS region name
  --statistics STATISTICS
                        statistics to compare (default: Average)

required arguments:
  --name NAME           metric name (e.g. CPUUtilization)
  --namespace NAMESPACE
                        metric namespace (e.g. EC2)
  --warning WARNING     warning threshold
  --critical CRITICAL   critical threshold

thresholds and ranges:
  Threshold ranges are in Nagios format:
  https://nagios-plugins.org/doc/guidelines.html#THRESHOLDFORMAT