Live Metrics shows how much memory and storage the platform is using right now, and whether each service is healthy. Use it to check the platform's state at a glance, or to find out when a service was down on a past day.
You see only the tabs your permissions allow. See Observability for the other tabs.
The top of the page shows two cards.
Each card has a ring that shows the percentage in use, a Capacity usage bar, and three figures:
A label in the corner of each card reads Normal, High or Critical as the resource fills up: Normal below 40%, High from 40% to under 70%, and Critical at 70% or more. The Capacity usage bar marks the 40% and 70% points on its scale.
Next to Used and Available, a small arrow shows how the figure moved since the reading before it. For Used, a rising value is amber and a falling value is green; for Available, a rising value is green and a falling value is amber. The arrow next to Total stays grey. An arrow appears only once the page has an earlier reading to compare with, and very small changes do not move it: less than a fraction of a percent for percentages, and less than about a megabyte for sizes.
The header reads Live metrics, next to a pulsing Live badge. Under it, the Refresh in counter shows the seconds until the next update. While the numbers reload, the counter changes to Updating….
Under Service health, each monitored service has one row:
The bars use four statuses, shown in the legend at the top right of the section:
To read one day:
Live Metrics has no settings. The resource numbers refresh every 5 seconds, the service health bars refresh once a minute, an open day's checks refresh about once a minute, and the history covers up to 90 days.
On wide screens, the Dashboard's right-hand rail shows the compact Memory and Storage cards with a health summary above them. The summary reads:
Hover the summary to list the services that are Down and the ones that are Operational. See The Dashboard.
To open Live Metrics, you need the View Dashboard permission. The Dashboard Viewer permission grants it, and so do the Observability Administrator and Observability Viewer presets. See Roles and permissions and the Permission reference.
If View Dashboard is the only observability permission you hold, the Observability item still appears in the sidebar, and Live Metrics is the only tab you see.
Every 5 seconds while the page is open. The Refresh in counter counts down the seconds to the next update, and the service health bars refresh once a minute.
Live Metrics shows current memory and storage figures, and each service's health for up to 90 days. System Overview shows how memory, CPU and storage were used over time, and the incidents that happened.
View Dashboard. The Dashboard Viewer permission grants it, and so do the Observability Administrator and Observability Viewer presets.
They show how full a resource is. Below 40% the label is Normal, from 40% to under 70% it is High, and at 70% or more it is Critical. The Capacity usage bar marks the 40% and 70% points.
The disk space in use on the primary volume. It shows Used, Available and Total.
Up to 90 days. Each service shows one bar per day. A service with less history shows fewer bars.
A grey day has the status No data, which means no health checks were recorded for it. There is nothing to show, so the day does not open.
Click the day's bar. The panel marks each 5-minute slot as Operational, Outage or No data, and the red badge next to the date shows the total time down for that day, for example 25m. Hover a slot to see its time.
The service is not healthy now. Open System Overview, find incidents for that service around the time it went down, and check its memory and CPU use and any restarts it shows.
The page could not get the resource figures. The message under the heading gives the reason, or reads "Unable to load live system metrics." if there is none. See Observability troubleshooting.
They compare the figure with the reading before it. For Used, a rising value shows amber and a falling value green; for Available, the other way round. Very small changes do not move the arrow, and the arrow next to Total stays grey.
It reads All systems operational, Partial outage with the number of services down, Major outage - all services down, or No systems monitored. Hover it to list the services that are down and the ones that are operational. See The Dashboard.