Monitoring Device

What the device monitoring page shows

The device monitoring page is a drill-down for one monitored network device — a switch, router, firewall, PDU, access point, or load balancer. It gathers everything about that device on one screen: live health gauges, time-series charts for traffic, CPU, memory and temperature, a per-port table, and the alerts raised against it.

How to reach it

Open Monitoring from the activity icon in the top header bar, switch to the Devices tab, and click any row. You can also arrive from the Overview tab by clicking a Top Talkers row or an entry in the Active Alerts panel. A Back to Monitoring link sits above the device name to return you to the list.

The header

The device name is the page heading. Underneath it runs a compact summary line:

• Management address — the address being polled.
• Device type — switch, router, firewall, and so on.
• Model — shown when the device reports one.
• Uptime — for example Up 12d 4h, shown when the device reports it.
• Last polled — the time of the most recent successful poll, or Never.

Header buttons

Poll Now

Queries the device immediately instead of waiting for the next five-minute cycle. The icon spins while it runs, then the gauges, charts and port table reload. Rates are the difference from the previous poll, so a poll fired within a couple of minutes of the last one produces a less meaningful rate than one taken on the normal cadence.

Thresholds

Opens the Device Thresholds window for this device, where you set the warning and critical limits that raise alerts. See below.

View Device

Opens the device's full record from Inventory in a window, so you can check or change its address, credentials, location and ports without leaving this page.

Health gauges

A row of four cards gives the current picture. It appears only once the device has reported health data, so a device added a moment ago shows nothing here until its first successful poll.

• CPU — current processor use as a percentage.
• Memory — current memory use as a percentage.
• Temperature — current temperature in degrees Celsius.
• Status — the hardware Fan and PSU readings, each coloured green for ok, amber for a warning, red for critical.

The first three cards carry a coloured bar under the number. It is green while the reading is comfortable, amber once it passes a warning level, and red past a critical one. Those levels are fixed for the gauges: CPU turns amber at 80% and red at 95%, memory amber at 85% and red at 95%, temperature amber at 70°C and red at 85°C. A reading of N/A means the device did not report that value on its last poll — the same goes for Fan and PSU, which only appear on hardware that exposes those sensors.

Note: these gauge colours are a reading aid and are separate from the alert limits you set with the Thresholds button. Changing a threshold changes which alerts are raised, not where these bars change colour.

Choosing a time range

A toggle above the charts sets the window for every time-series chart on the page: 1 Hour, 6 Hours, 24 Hours, or 7 Days. It starts on 24 Hours. Use 1 Hour to zoom into a recent spike, 7 Days to look at a trend.

The charts

Aggregate Traffic

An area chart combining the inbound (green) and outbound (blue) rate across all of the device's ports over the chosen window. Hovering a point shows the rate in bits per second.

CPU & Memory

A line chart plotting CPU (blue) and memory (purple) use as percentages on a 0–100 scale, so you can see sustained load or spikes that move together.

Temperature

A line chart of temperature in degrees Celsius over the same window.

Note: each chart only appears once there is history to draw. If one is missing, the device either has not collected data for that window yet or does not report that measurement at all. Pick a shorter range, or come back after a few more polling cycles.

Ports table

When the device exposes ports, a table lists them. The heading shows how many are up out of the total — for example Ports (22/48 up).

Ports table columns

Column	What it shows
Port	The port name, or a numbered label when the device reports none.
Status	A coloured dot and word — green up, red down.
Speed	The link speed in Mbps. Hidden on narrow screens.
In	Current inbound rate, in bits per second.
Out	Current outbound rate, in bits per second.
Utilization	A bar plus a percentage for the busier of the two directions — green below 70%, amber from 70%, red from 90%. Hidden on smaller screens.

Alerts

The Alerts panel lists the recent alerts belonging to this device, with the heading showing how many are currently active — for example Alerts (2 active). Each row carries:

• A severity badge — critical, warning, or info.
• The alert type, with the port name beside it when the alert is about one port.
• A short message describing the condition, including the reading and the limit it crossed.
• The timestamp of when it was raised.

An alert that is still active and not yet acknowledged shows an Acknowledge button on the right; clicking it marks it as seen while leaving it in the list. An alert that has cleared shows a green Resolved badge instead. When there is nothing to show, the panel displays a green tick with No alerts — This device has no alerts.

Tip: this panel shows the most recent alerts for this device. For the whole estate, filters, and bulk acknowledge, resolve or delete, use the Alerts tab on Monitoring.

Setting thresholds for this device

Click Thresholds in the header to open Device Thresholds. The window explains itself: Set warning and critical thresholds for alert triggers. Each row has an enable checkbox, a W (warning) value, a C (critical) value, and its unit. The rows are grouped:

Device threshold settings

Section	Threshold	Unit
Traffic	Inbound Utilization	%
	Outbound Utilization	%
	Error Rate	/min
	Discard Rate	/min
Health	CPU Usage	%
	Memory Usage	%
	Temperature	°C

To set a limit, tick its checkbox, type the warning and critical numbers, and click Save. Untick a row to stop alerting on that measurement for this device; its value boxes grey out. Cancel or the close icon discards your changes.

Note: the limits you set here apply to this device. Set them on each device you want alerting for — a device with no limits recorded raises no threshold alerts. Traffic thresholds are checked per port, so one busy port on a quiet switch is enough to raise an alert.

Common tasks

1. Get fresh numbers: click Poll Now and wait for the gauges and charts to reload.
2. Investigate a spike: set the range to 1 Hour and read the Aggregate Traffic and CPU & Memory charts together.
3. Quiet a known issue: click Acknowledge on the alert so it stops reading as needing attention, without closing it.
4. Tune sensitivity: open Thresholds and adjust the warning and critical values.
5. Check a single link: read the Ports table to see which ports are up, how fast they are, and how loaded each one is.
6. Fix a device that will not poll: click View Device, correct its management address or credentials, close the window, and click Poll Now.

Related articles

Monitoring · Inventory · DDoS Protection