This Lambda function has been created as AWS CodePipeline currently doesn't allow to send notifications to report the status of the pipeline. Hopefully this feature will be available in the future.
The Lambda function is executed again and again (using the continuation token), until the action it is monitoring succeded or failed. The fact that it is monitoring makes the entire pipeline slower (about 1 minute per monitored action) compared to a Lambda script that would just send a simple notification.
How to use
Lambda
Create a Lambda function using the code provided and the Python 3.6 runtime. Attach an IAM role that has at least the AWSLambdaBasicExecutionRole policy as well as this CodePipeline permissions:
- "codepipeline:PutJobSuccessResult"
- "codepipeline:PutJobFailureResult"
- "codepipeline:Get*"
Slack
Create a Slack incoming webhook.
Lambda environment variable name: kmsEncryptedHookUrl
Get the webhook url (without the https://) and add it to the Lambda function as an encrypted environment variable.
Lambda environment variable name: slackChannel
Also add the Slack channel name you want to send the notifications to as an environment variable (no encryption).
CodePipeline
Add an Invoke action in parallel to the action you want to monitor. Choose AWS Lambda as the Provider and select your Lambda function. Enter the name of the action you want to monitor as User Parameters.