Data Gathering

[MStahl15]

Greetings!

The company I work for deals with PHI/PII (Protected Health Information/Personally Identifiable Information) and I'm curious as to the data that Lighthouse gathers on its end, if any?

For obvious reasons, dealing with peoples PHI/PII, this information cannot be exposed or gathered by un-approved or un-contracted third parties so I'm hesitant on implementing Lighthouse into the applicable repos.

Any insight to this will be much appreciated!!

[connorjclark]

If you're running Lighthouse yourself locally (CLI, DevTools, ... NOT the chrome extension), you'll be happy to know we collect nothing except for the opt-in Sentry gathering in the CLI (which will only send us stack traces and URLs).

[MStahl15]

What about runs that are not local? Looking to have this integrated in our CircleCI pipeline running against multiple environments, including production via a CRON job.

If the answer is the same, then awesome - I'll move forward with implementation. If not, would you mind explaining what exactly is gathered in this scenario, or point me to the proper documentation?

[brendankenny]

If you're using the CLI or node module, those count as local runs too (just local to the CI machine) and Lighthouse logs no information by default. A non-local run would be something like PSI, where Lighthouse is hosted and run for you and is covered by the usual Google privacy policy (or whoever is hosting the service if it's somewhere else).

[brendankenny]

docs for the error logging @connorjclark mentioned: https://github.com/GoogleChrome/lighthouse/blob/master/docs/error-reporting.md

but as he said, it's opt-in for the CLI only, and there won't even be an opportunity to opt-in because the prompt is disabled when process.env.CI is set to true (as it is in CircleCI)