System status and platform health

2 min · Screen: /admin/system · Needs: Permissions.System.Health

The degraded banner

When one of the platform’s own health checks fails, every page shows a red banner: The platform has been degraded since …. Backups may not run or be stored as they should. Click System status in the banner to see which check fails. The banner and its alert clear themselves once every check passes.

Admin › System status on a degraded platform: the storage check fails

On this screen:

• The degraded banner, on every page
• The platform’s own health checks
• Outbound heartbeat
• Version and instance

What System status shows

Section	Tells you
Health checks	Each check with its state and the time it was read.
Outbound heartbeat	Whether a heartbeat is set up, and when it last pinged.
Platform	Version, instance id and start time.
Database	Provider, server version, whether it is reachable, migrations applied and pending.
Jobs since this restart	Whether start-up reconciliation finished, job servers and queues.

Note

The job engine keeps nothing across a restart: schedules, queued runs and retries live in the platform database and are rebuilt at start. Job figures here count from the last start only.

Platform health in three steps

1. Read the failing check’s message on Admin › System status

2. Fix the cause

   Usually storage (a target unreachable or full), the database, or the master key. See Fix common problems.

3. Refresh - the check, banner and alert clear when it passes

Tip

Set up a Heartbeat channel so you hear about it even when the platform is completely down. See Notification channels and rules.

Related

• Alerts and the notification centre
• Notification channels and rules
• Fix common problems
• Back up the platform itself