Devices

This page is also in the Groundskeeper wiki: https://github.com/rsjsoftwareltd/groundskeeper/wiki/Devices

The Devices page lists every machine and piece of network kit that Groundskeeper knows about, in one table. Computers that run the RSJ Agent sit beside the switches, access points, servers, UPSes and printers that your connectors watch.

Open it from the Devices item in the sidebar, or at https://<server-address>:8443/devices. Admin and technician accounts can see it. An SLT account cannot.

The page loads itself again every 30 seconds. You do not need to reload it.

Where the rows come from #

Each row comes from one of these sources. The Type column says which.

Type shown	Where it comes from
Workstation, Server, DC	The RSJ Agent on the machine. A machine that reports itself as a domain controller is shown as DC.
WMI	A Windows server watched over WMI by the Windows Servers connector, with no agent installed. If the agent also reports the same server, only the agent's row is shown.
AP, Switch, Network	Cisco Meraki devices (wireless, switch, and other network products).
AP, Switch, Network	Aruba Central devices (access points, switches and gateways) appear here too, under the same filters, once the Aruba Central connector has polled.
Switch	HP and Aruba switches read over SNMP also appear here, under the same filter.
iDRAC	Dell iDRAC servers.
UPS	UPSes read over SNMP.
Printer	Printers and MFDs read over SNMP.

A source only fills in once its connector is set up and has polled. See Connectors.

If part of the list could not be loaded, a red notice above the table names the sources that failed and says the list may be incomplete. An empty table with no notice means nothing has reported yet. In that case the page points you to Agent Settings to deploy the agent.

The summary chips #

Across the top are five chips. Click one to filter the table to that group. Click total to clear it.

total -- every device the page has loaded.
online -- reachable and healthy.
offline -- not reachable.
stale (>24h) -- an agent machine that has not checked in for more than 24 hours.
warning -- devices that have a live, unacknowledged warning on the feed from the last 7 days. The number on a row is how many distinct warnings it has. Acknowledged or snoozed warnings are not counted, and neither are warnings that a later event has replaced.

How "online" is decided depends on the kind of device:

Agent machines: online if the agent has checked in within the offline time set under Agent Settings -> Monitoring policy -> Device offline (15 minutes by default). Longer than that is offline, and longer than 24 hours is stale. This is the same rule the dashboard uses.
Meraki devices: Meraki's own status. A Meraki device that is alerting or dormant is counted in neither online nor offline.
iDRAC, WMI servers, UPSes and printers: healthy counts as online. Unreachable counts as offline. A warning or critical state counts as neither, because the device is still answering.
Aruba Central devices: online, offline or alerting, as Aruba Central reports them. Alerting is counted in neither online nor offline.
HP and Aruba switches: reachable is online, unreachable is offline.

The small cog beside the chips opens a list of device types. Untick a type to keep it out of the offline and stale counts and filters. This is useful when many laptops are off overnight and you want to see an offline server instead. It changes nothing else: the table, the other chips and the alerts are not affected, and the choice is not remembered after you reload the page. The list offers workstations, servers, domain controllers, access points, switches, other network, iDRAC, agentless Windows servers, UPSes and printers.

Finding a device #

The search box matches the hostname, the signed-in user (for Meraki devices, the organisation), the operating system or model, and tags.
The type buttons under it filter by kind: Workstations, Servers (this includes WMI servers), Domain controllers, Access points, Switches, UPS, Printers, Other network, and iDRAC.
The tag row filters by your own tags. See below.

Search, type, chip and tag filters combine. The count at the right of the search bar reads "N devices", or "N of M" when a filter is on.

Export CSV downloads exactly the rows the table is showing now. The columns are hostname, type, status, live warnings, OS or model, user or organisation, CPU %, RAM %, disk C: %, tags, last seen, model, serial and address. It opens correctly in Excel, and a cell that starts with =, +, - or @ is prefixed so Excel does not run it as a formula.

Reading the table #

Column	What it shows
Hostname	The device's name. Click the row to open its drawer.
Type	The badge from the table above.
Status	Online, Offline, Stale, Alerting, Dormant, OK, Warning, Critical or Unreachable, depending on the source. A blue maintenance pill appears if the device is inside an active maintenance window. The device is still shown as offline: the pill explains why. An amber number is the live warning count.
OS	The operating system for agent machines. For network kit, iDRAC and printers, the model. For some switches, hover to see the full description from the switch.
User	The signed-in user for agent machines. For Meraki devices, the organisation. For iDRAC, the Dell service tag.
CPU, RAM, Disk C:	A bar and a percentage. Amber from 80% CPU, 85% RAM and 90% disk. Red from 90% CPU, 90% RAM and 95% disk. A dash means the source does not report it.
Tags	Your tags and sections on this device.
Last seen	How long ago the device last reported. An amber Reboot pill means Windows is waiting for a restart.

The device drawer #

Click a row (or focus it and press Enter) to open a panel on the right with more detail. Press Escape or click outside it to close it. What it shows depends on the type of device.

Every drawer starts with the status, when it was last seen, any live warnings by title, and the Tags box.

Agent machines (workstations, servers, DCs) also show the sections below. A section only appears when the agent has reported that information.

Health -- CPU, RAM and disk C: free space.
Volumes -- each drive with its percentage used and free space.
Security -- BitLocker, Defender, and whether the antivirus signatures are up to date.
Updates -- pending reboot and the date of the last update.
Windows 11 readiness, with blockers and anything that could not be measured.
Services -- automatic-start services that are not running.
Hyper-V -- running and total VMs and replication errors, on hosts that report it.
Inventory -- maker, model, serial, RAM and the number of installed programs.
Agent -- the agent version, the last check-in time, and a count of collectors that are failing, if any. On a machine running the RSJ Agent host it also shows the host version, with a note if an update is available or has failed, and whether the Groundskeeper module is current. It says the machine is observe-only with a count of read-only modules.
Installed modules -- every module the host runs, each marked observe-only or acts on device.
Module activity -- recent module install and remove requests, and whether each is pending, done or failed. This only appears once there has been some.

Meraki devices show model, product type, serial, MAC, network ID, organisation, any active Meraki alerts, and, where reported, the neighbours the device sees over LLDP.

Aruba Central devices show model, device type, serial, MAC, group and site.

HP and Aruba switches show health, uptime, availability over 30 days (hover for the outages, time down and reboots behind the figure), port count, firmware, a grid of monitored ports (green up, red down, grey otherwise; hover a port for its name and status), the LLDP neighbours, and a table of what is on each port: whether it is an uplink or a device port, and what was last seen at the other end. Only ports the Switches connector monitors appear.

iDRAC servers show host, model, service tag, power state, and whether the iDRAC is reachable.

WMI servers show health, CPU, RAM, uptime and last reboot, with a note that they are watched without an agent and that deploying the agent gives more detail.

UPSes show whether the UPS is on mains or on battery, battery %, runtime and load.

Printers show status, model, and toner or ink levels and paper tray levels where the printer reports them. The bars are plain, not coloured, because the low-supply thresholds belong to the printer connector and its alerts.

Tags and sections #

Tags are your own labels for grouping devices, such as "Site: Main Building". They are stored on the Groundskeeper server, so everyone sees the same tags.

Add a tag: open the device's drawer, type in the Tags box and press Enter or click Add tag. Tags are matched by device name regardless of capitals.
Remove a tag: click the x on the tag in the drawer.
Filter by a tag: click it in the tag row above the table. Click it again, or click clear, to stop filtering.

A section is a tag with a friendly name and a colour. Click the + on a plain tag (in the tag row or in a drawer) to turn it into a section: enter a display name, pick a colour and save. The pencil on a section edits it, and Remove label takes the name and colour off. Removing the label does not touch any device's tags: the tag just goes back to a plain chip.

Admin and technician accounts can both add and remove tags and sections. Tags and sections are only for organising the page. They do not change alerts.

What this page does not do #

It only shows what the agent and connectors have reported. It cannot restart a service, install an update or change anything on a device. Groundskeeper observes and does not act.