Application: govuk-mirror

A simple Go program used by GOV.UK to populate mirrors hosted by AWS S3 and GCP Storage

• GitHub: govuk-mirror
• Ownership: #govuk-platform-engineering owns the repo. #govuk-platform-support receives automated alerts for this repo.
• Category: Utilities

README

A concurrent crawler and site downloader to make a local copy of a website. This is used by GOV.UK to populate mirrors hosted by AWS S3 and GCP Storage.

Usage

Configuration is handled through environment variables as listed below:

• SITE: Specifies the starting URL for the crawler.
  • Example: SITE=https://www-origin.publishing.service.gov.uk

• ALLOWED_DOMAINS: A comma-separated list of hostnames permitted to be crawled.
  • Example: ALLOWED_DOMAINS=domain1.com,domain2.com
• USER_AGENT: Customizes the user agent for requests. Defaults to govuk-mirror-bot if not specified.
  • Example: USER_AGENT=custom-user-agent
• HEADERS: Provides custom headers for requests.
  • Example: HEADERS=Rate-Limit-Token:ABC123,X-Header:X-Value
• CONCURRENCY: Controls the number of concurrent requests, useful for controlling request rate.
  • Example: CONCURRENCY=10
• URL_RULES: A comma-separated list of regex patterns matching URLs that the crawler should crawl. All other URLs will be avoided.
  • Example: URL_RULES=https://www-origin.publishing.service.gov.uk/.
• DISALLOWED_URL_RULES: A comma-separated list of regex patterns matching URLs that the crawler should avoid.
  • Example: DISALLOWED_URL_RULES=/search/.,/government/.*.atom
• SKIP_VALIDATION: Skip domain accessibility validation before crawling. Useful for offline testing.
  • Example: SKIP_VALIDATION=true

How to deploy

This needs manual deployment to staging and production. Once the Release GitHub Action has run select the Run workflow option from the Deploy GitHub action. Then enter the latest tag number and the environment to deploy to.