Trigger Web Page Requests with AWS Lambda

Use an AWS Lambda to send an HTTP request to a destination.

Create a Lambda

First, check out the repository and use this command to create a .zip file to upload to AWS Lambda.

npm run package-for-deploy

While creating the Lambda, you should probably go to the Advanced settings and change the maximum timeout. Web pages sometimes need more than the 3 seconds that are allocated by default.

Triggering the Lambda

Use AWS CloudWatch and create a new event rule. Select the "cron" option and describe the schedule you want for how often a web page will get hit. For example, every day at 3 AM UTC would look like this:

0 3 * * ? *

While customizing the event, select "Configure input" and select "Constant (JSON text)". The JSON is where you configure this Lambda to issue a web page request. Here is an example:

{
    "dryRun": true,
    "method": "GET",
    "password": "There is no spoon.",
    "urlTemplate": "http://example.com/hitcounter?{YYYY}-{MM}-{DD}",
    "username": "Neo"
}

This will issue a "GET" request to the example.com address. The urlTemplate can inject values from the current date/time into the URL. See date.format for more details.

Enabling dryRun will log messages to output but will not issue the web page request.

If you want multiple web pages retrieved or if you want a more complex schedule, simply set up more CloudWatch triggers with custom data.

Logging

The script writes output to the logs. Make sure to set up a policy to clean up the logs after an amount of time. The entire response body is also logged, which could be large.

License

This is released under a modified MIT-style license.