cloudwatch-mon-scripts-python

Linux monitoring scripts for the AWS CloudWatch Service.

Couple of additional features (compared the original AWS monitoring scripts):

• Memory monitoring incl. buffers
• Load monitoring (overall and per CPU core)
• Monitoring of disk inode usage
• Process monitoring
• Fewer dependencies
• Simpler installation

Requirements

• Python 2 (>= 2.6) or Python 3 (>= 3.3)
• Boto >= 2.33.0

Installation

Optionally create a virtual environment and activate it. Then just run pip install cloudwatchmon. Install the scripts in /usr/local/bin folder.

For script usage, run:

mon-put-instance-stats.py --help

Examples

To perform a simple test run without posting data to Amazon CloudWatch:

mon-put-instance-stats.py --mem-util --verify --verbose

Report memory and disk space utilization to Amazon CloudWatch:

mon-put-instance-stats.py --mem-util --disk-space-util --disk-path=/

To get utilization statistics for the last 12 hours:

mon-get-instance-stats.py --recent-hours=12

Configuration

To allow an EC2 instance to read and post metric data to Amazon CloudWatch, this IAM policy is required:

{
  "Statement": [
    {
      "Action": [
        "cloudwatch:ListMetrics",
        "cloudwatch:GetMetricStatistics",
        "cloudwatch:PutMetricData",
        "autoscaling:DescribeAutoScalingInstances"
      ],
      "Effect": "Allow",
      "Resource": "*"
    }
  ]
}

If the policy is configured via an IAM role that is assigned to the EC2 server this script runs on, you're done.

Otherwise you can configure the policy for a user account and export the credentials before running the script:

export AWS_ACCESS_KEY_ID=[Your AWS Access Key ID]
export AWS_SECRET_ACCESS_KEY=[Your AWS Secret Access Key]

Third option is to create a ~/.boto file with this content:

[Credentials]
aws_access_key_id = Your AWS Access Key ID
aws_secret_access_key = Your AWS Secret Access Key