Lighthouse as a service is designed to be used to
- AWS CLI Installed
- Permissions to create all required resources (see below for resources that are created)
Clone this repo then run the following command.
npm run deploy
This will create the following stack, clone this repo onto spot instances and run the worker script to audit pages.
Running ./scripts/deploy.sh will do the following.
- Creates a new VPC in
us-east-1. - Creates two subnets in different availability zones within the VPC.
- Creates and attaches an internet gateway to the VPC.
- Creates and attaches route table with a route to the internet gateway to the subnets allowing internet access from the two subnets.
- Creates policies to be assumed by a launch template.
- Creates a launch template defining an instance type of
m5.largewith anAmazon Linux 2 AMIimage, assuming a policy that enables it to interact with autoscaling, sqs and lambda. The launch template is also provided with a start up script that installs required tools to run lighthouse tests. - Creates an SQS Queue for audits.
- Creates an autoscaling group of spot instances with a desired capacity of
1and min capcity of0and a max capacity of1. - Creates an autoscaling policy acting upon the autoscaling group where the desired state is the Audit Queue to have 0 messages.
Updates the Cloudformation Stack
Used as part of the instance start up script, it installs dependencies and starts the worker node script.
Add to the SQS Queue the following.
{
"domain": "https://hubspot.com",
"callback": "<your_application_url>"
}Once the audit is complete, a POST request will be sent to your callback url with the url of your report. Reports are kept for 7 days before being deleted.