aws-codepipeline-events

Monitor your AWS CodePipelines with CloudWatch, Datadog and Slack.

This project was created as an internal tool to get information about succeeded and failed steps in CodePipelines deployed in almost all regions available in Amazon Web Services. As CloudWatch does not provide any metrics for this service, we had to create something on our own. Initial idea was to subscribe to Cloudwatch Events and push everything to Datadog Events, but then we have added integration for Slack and a scheduled lambda which pushes custom metric to CloudWatch.

Architecture

Features

• Datadog Events with information about changed pipeline, step and action statuses
• Slack notification about changed pipeline status
• Cloudwatch custom metric with pipeline status (0 - succeeded, 1 - in progress, 2 - failed)
• Receives events from all regions.

Note: this solution is not easily customizable and probably you will need to modify serverless.yml or handler.ts to remove features you do not need.

Requirements

• Serverless Framework
• Datadog API key stored as /datadog/DD_API_KEY in AWS SSM Paramter Store
• Slack Webhook URL stored as /slack/SLACK_WEBHOOK_URL in AWS SSM Parameter Store

Deployment

sls deploy --region eu-west-1 --stage prod

This will create Lambda functions and SQS queue which is needed for next step.

for i in $(aws ec2 describe-regions --query "Regions[].{Name:RegionName}" --output text); do aws cloudformation deploy --stack-name codepipeline-events --template-file cfn/subscriber.yml --region $i --parameter-overrides "PUT SQS ARN HERE" --capabilities CAPABILITY_IAM; done

Or you can use StackSets 😉

Authors

• Jakub Wozniak, Nordcloud 🇵🇱