Skip to main content

Health

Platform → Health (/platform/health), staff only. The first stop when something seems wrong, and the page to open before saying the service is fine.

The service​

Every hosted address with its uptime over 24 hours, 7 days and 30 days; each service's traffic over 30 days with the share of requests answered with a server error and its median and 95th-percentile response times; the last database backup; the last verified restore; and when each nightly job last ran and whether it succeeded. These come from the public status feed (the setting STATUS_FEED_URL), the same source as the public Status and Traffic pages.

Activity​

Organisations, people, people active in the last 24 hours and 7 days, annotations in the last 24 hours and 7 days, projects and tasks.

Inside the platform​

  • Database: the engine and how long a trivial query takes, red above 250 ms or when it does not answer.
  • Email: the backend in use, red when it is one that keeps or drops messages (console, file, dummy).
  • Storage connections: how many across every project, and each one whose last sync failed, with the provider's error.
  • Exports in progress and failed in the last 7 days.
  • Rows added by installed extensions.

For extension authors​

An extension answers the question health_rows(organization) with rows {"label", "state", "text"} (state one of good, bad, none): with an organisation for its Health page, with None for the platform's. See Extensions.