Who can use this

System Metrics

System Metrics is your live operational dashboard for the server that runs Grengin. It shows how hard the host machine is working, how healthy the database is, and what limits the container is operating under, all in one place. Use it to spot infrastructure pressure early and to confirm the backend is up and responding.

Grengin system metrics

Who can use this

To open System Metrics you need the View Analytics permission, granted organization-wide. A department-scoped analytics grant will not unlock this page, it must apply across your whole organization.

This page is reached directly from the Control Hub and does not appear as a regular sidebar item.

If you also hold the Maintain System permission, this page gains an extra Maintenance tab alongside the Metrics tab. That permission is only needed for maintenance work, not for viewing metrics. (See System Maintenance.)

Reading the status banner

A banner runs across the top of the page and gives you an at-a-glance health summary:

  • Version of the application currently running.
  • Health indicator showing All Systems Operational with a green dot when everything is healthy, or the reported status with a red dot when something is off.
  • Generated at timestamp telling you exactly when this snapshot of the metrics was taken.

If the version cannot be read at the moment, you will briefly see a Loading… placeholder in its place.

Checking machine health

The Machine section shows resource usage for the host server as a grid of stat cards:

  • CPU Usage with a progress bar. The bar shifts to a warning color above 70% and a critical color above 90%.
  • Memory Usage as a used-of-total percentage, with the raw amounts shown underneath.
  • Load Average across the last 1, 5, and 15 minutes.
  • Uptime for the host, shown in days, hours, and minutes.
  • Swap Usage as a used-of-total percentage with amounts. This card only appears when the host actually has swap configured.
  • Disks listing each mount point with its used and total space and a percentage. The bar warns above 80% and turns critical above 95%. This section only appears when disks are reported.

Checking database health

The Database section reports on the platform’s PostgreSQL database:

  • Connections, the total number open, with badges for how many are active versus idle.
  • Latency, the database round-trip time in milliseconds.
  • Size of the database.
  • Cache Hit Ratio as a percentage, shown in green when above 95%, which signals the database is serving most reads from memory.
  • Transactions, showing commits and rollbacks. Rollbacks are flagged in a warning color so they stand out.
  • Tuples, breaking down rows fetched, inserted, updated, deleted, and returned.

Checking container limits

The Container section shows the limits the application is running under:

  • Inside Container, a Yes or No badge.
  • Cgroup Version, or a dash when not applicable.
  • CPU Quota in cores and CPU Usage in seconds.
  • Container Memory Usage with a bar showing usage as a percentage of the limit. The percentage detail only appears when a memory limit is set.
  • Container Memory Limit and Container Memory Available.

Refreshing the data

The numbers reflect a single snapshot in time, taken when the page loaded. To update them:

  1. Click Refresh in the page header to pull a fresh snapshot immediately. The button is disabled while a refresh is already in progress.
  2. Turn on the Auto-refresh toggle to have the page reload the metrics automatically every 15 seconds. Turn it off to stop.

If a refresh fails to load, you will see an error card with a Retry button. Click it to try again.

Good to know

  • All charts here are live progress bars driven by the current snapshot. There are no historical or time-series graphs, so this view tells you how things look right now, not over time.
  • The version and health banner is read once when you open the page and is not re-pulled on later refreshes, so treat it as the state from when you arrived.
  • Hosts without swap, without disk reporting, or without container memory limits will simply show fewer cards. Missing cards mean that data does not apply to your environment, not that something is broken.
  • While you are loading the page you will see a centered spinner; metrics appear once the snapshot arrives.
  • The Maintenance tab only appears if you hold the Maintain System permission, and it is a configuration surface rather than a metrics one. It is covered in System Maintenance.

Part of the Control Hub Guide.