Health Dashboard

Purpose

• Land on one account-wide health summary immediately after sign-in.
• Check the state of every resource category you can access at a glance.
• Spot open alerts, failed backups, and quota pressure without leaving the page.
• Identify the heaviest-consuming VMs and the fullest disks.

Prerequisites

The dashboard requires any one of its metric read permissions; each section renders only if you hold its own permission.

• Resource health tiles need: compute.vms.read, networking.firewalls.read, networking.vnets.read.
• Disk inventory needs: compute.disks.read.
• Clusters, hosts, storage, and patch status (root account only) need: platform.clusters.managed.read.
• Open alerts by severity (root account only) need: platform.alerts.read.
• Backup failures need: dataprotection.backups.jobs.read.
• Quota headroom needs: organisation.quotas.read.
• Top consumers need: compute.monitoring.read (VM usage also accepts platform.globalvms.read).

View the dashboard

The dashboard is read-only and refreshes on its own.

1. Open Home; the Health Dashboard is the landing page shown after sign-in (also at /dashboard).
2. Read the resource tiles: each category shows the healthy count against down, degraded, and unknown.
3. Review the Alerts and Backups panels for open problems and failed backup runs.
4. On a root account, check the Platform section for per-cluster host, storage, and patch status.
5. Check Quota headroom for vCPU, memory, and storage remaining.
6. Scan Top consumers for the busiest VMs by 24-hour peak CPU and the fullest disks.
7. Expected result: a current, permission-scoped health summary that refreshes automatically about every 60 seconds.

Best Practices

• Treat the dashboard as a starting point; act on any issue from its owning section (Alerts, Backups, Quotas).
• Use it as your first check before a demo or after a change.
• Remember counts are scoped to the accounts you can see; child-account resources appear only on parent or root accounts.
• Note that Top VMs and Top disks cover the active account only and exclude resources with no usage telemetry.