A command-line tool that retrieves a sitemap.xml from a site and audits every page contained within using Lighthouse+Headless Chrome. The page metrics are then reported to Application Insights for analysis.
Features:
- Uses a pool of chromium instances to minimize chrome process startup overhead
- Supports concurrent lighthouse processing in isolated processes (on suitably beefy hardware)
- Auto-retry of failed urls
- Auto-timeout of audit sessions, both from the chrome process and lighthouse worker perspectives
- Stores lighthouse reports of all runs either locally or in azure storage - can drag reports on to https://googlechrome.github.io/lighthouse/viewer/ to see the visual report
- Provides visual monitoring progress
| Option | Description | required |
|---|---|---|
--url |
The absolute url to a sitemap.xml (required) | [string] |
-c |
The number of concurrent workers to use (optional) | [default: 1] |
-i |
The number of iterations of all urls to run (optional) | [default: 1] |
--config-path |
Path to a lighthouse configuration file (optional) | [string] |
--output-path |
Indicates the path which report files will be saved (optional) | [default: './output/] |
--resume |
Indicates if a previous run will be resumed by examining existance of output files (optional) | [default: false] |
--container-name |
Indicates an existing container name to use (required when using the --output-Path azure-storage option) |
[string] |
--tag |
Adds a custom dimension named 'tag' to each AI metric (optional) | [default: start date/time] |
An
APPLICATIONINSIGHTS_CONNECTION_STRINGenvironment variable needs to be set with the connection string of the intended AI instance.
When using the
-output-path azure-storageoption,AZURE_STORAGE_ACCOUNT_ACCESS_KEYandAZURE_STORAGE_ACCOUNT_NAMEenvironment variables need to be supplied, as well as a previously created--container-name
Important: When specifying the concurrency option, carefully monitor CPU/Memory/Network of the host. It's very easy to overload a host, thus skewing the results.
Audit all links at a url with a maximum of 8 simultaneous chrome processes:
yarn start --url https://new.site.com/sitemap.xml -c 8
Specify a number of iterations to crawl each page.
yarn start --url https://new.site.com/sitemap.xml -c 6 -i 3
Use a config file to specify the lighthouse configuration
yarn start --url https://new.site.com/sitemap.xml -c 8 --config-path ./lighthouse.config.json
Resume processing the specified sitemap by using output report files to determine if a url should be processed
yarn start --url https://new.site.com/sitemap.xml -c 6 --resume
Specify a different output path
yarn start --url https://new.site.com/sitemap.xml -c 6 --output-path ./reports
Store Reports in Azure Blob Storage in a container named 'reports'
yarn start --url https://new.site.com/sitemap.xml -c 6 --output-path azure-storage --container-name reports