ZenPlus · Device Tracker guide

Device Tracker: complete operator guide

Set up endpoint, switch-port and user-activity tracking, verify the evidence, and investigate changes with a repeatable workflow.

Applies to ZenPlus 1.24.13Updated 2026-10-11Audience Network operations & administratorsReading time ~12 min
Start with verified evidence

This guide describes the current appliance. Availability, vendor features and access depend on your device firmware and Windows policies. Screenshots use the development UI with fictional addresses, users and device names; they contain no customer inventory.

01 What Device Tracker can establish #

Device Tracker joins switch forwarding observations, gateway address observations and optional domain-controller authentication events. A MAC identifies an observed network interface; an IP address, port or username is evidence associated with it at a time. Coverage depends on the devices and contexts you collect.

Endpoint inventory: current IP evidence, observed switch port, VLAN, user activity and last-seen time.
Endpoint inventory: current IP evidence, observed switch port, VLAN, user activity and last-seen time. Development UI · fictional documentation data. Open image for full size.
QuestionWhere to workEvidence to check
Where was an endpoint observed?Endpoints and endpoint detailsSearch MAC / IP / hostname, then check the attachment label and observation time.
Which endpoints are on this port?Switch PortsSelect a switch and click the endpoint count; the resulting list keeps the switch and port filter.
Who authenticated from that address?User LoginsEvent time and address history; shared addresses may remain unassigned.
What changed?ActivityNew, moved, watch and rogue observations; match the event to its collection time.
What is approved?Watch ListsYour enabled allow / watch / ignore rules; rogue is a policy result.

02 First-time deployment checklist #

Plan the collection scope

Identify access switches, VLAN-routing gateways, controller-managed switches and optional DCs. Record model, firmware, management address and relevant VLAN / VDOM / VRF. Use the setup guide to provision access.

Prove read access

Test interface and forwarding tables with the same account used by the appliance. Test a relevant gateway neighbour table. A successful interface query alone does not prove MAC-table access.

Enable collection

Add monitored devices and their credentials; open Device Tracker → Settings, enable UDT and select overrides only where necessary. Start with the default 300-second interval.

Validate one known endpoint

Generate ordinary network traffic from an approved test workstation. Allow two polling cycles, then compare its MAC, address, VLAN and expected switch port with source-device read output.

Add optional Microsoft activity

Follow the dedicated Microsoft account guide. Use Poll now, then check controller status and event-to-endpoint evidence.

Apply policy and alerts

Review classification and allow rules before enabling rogue or watch alerts. Use an approved test endpoint and notification channel; do not disable production ports to test discovery.

Acceptance target

A known test endpoint has fresh, correctly scoped MAC/port observations; its IP reporter is visible; any user association has a matching event-time address. Missing data remains unknown rather than becoming an invented attachment.

03 The seven workspace tabs #

TabOperator task
EndpointsSearch inventory and open attachment, address and event history.
Switch PortsView direct and controller-managed switch inventory, status and observed endpoints per port.
User LoginsInspect authentication activity and correlated endpoints. Requires user-data permission.
ClassificationAssign meaningful endpoint types with automatic, rule or manual classification.
Watch ListsManage allow, watch and ignore policies.
ActivityInvestigate collection-derived changes over the selected time range.
SettingsSet polling intervals, credential overrides and domain controllers.

The current workspace is titled Device Tracker, at /udt. Screen labels may differ on older releases.

04 How observations become a tracked endpoint #

SourceContributesDoes not prove
Access-switch forwarding tableMAC → bridge port → interface; VLAN when available.The IP address or authenticated user by itself.
Gateway ARP / neighbour tableIP → MAC, reporter and routed interface.A physical access-switch port.
FortiGate switch-controller inventory + diagnosticsManaged-switch identities, local ports, link state and observed MAC locations.Direct endpoint attachment when access/uplink classification is unconfirmed.
Selected DC Security logsSuccessful authentication activity with event time and client address when present.Continuous desktop ownership or a reliable mapping behind a shared/proxy address.

SNMP is the primary source. Enabled SSH collection can supplement missing gateway or managed-switch evidence using fixed read commands. New or replacement address evidence is confirmed over repeated observations; an overlapping address remains visible as a conflict.

05 Read location and freshness correctly #

Label / signalInterpretation
DirectObserved on a port classified as access. Check topology: a phone, AP, hypervisor or unmanaged bridge can still sit behind it.
IndirectAn uplink / trunk or an observation that is not established as a direct attachment.
Observed port / unconfirmedController-reported MAC location. Useful evidence, but not a confirmed endpoint access port.
Gateway evidenceAn address reporter without a confirmed switch attachment.
Active nowSeen in the last 15 minutes; not a separate reachability test or a promise that the endpoint is powered on.
Unknown link stateThe source did not provide a supported operational state. Do not count it as available capacity.

A forwarding table can retain an old MAC after a move. Collection is periodic; attachment gaps of 15 minutes expire current state while retaining historical periods. Review all sightings and their timestamps before deciding a port is wrong.

06 Retention and polling performance #

Default UDT housekeeping retains closed attachment periods, address history, activity and user login history for 90 days. Endpoint identities and daily capacity records are retained separately. Check the appliance Storage page and your approved retention policy; this guide does not promise indefinite historical data.

Begin at 300 seconds. Shorten intervals for selected sources after measuring collection duration, timeout rates and Windows event volume. Keep the effective interval below the 15-minute active window if you need continuous active-state coverage. A very long interval can make a healthy endpoint appear inactive between polls. Page and filter large inventories instead of repeatedly loading every endpoint.

07 Permission boundaries #

CapabilityMinimum permissionNot required
Endpoint and port discoverySNMP read view on the relevant interface, forwarding and neighbour tables; appliance-source ACL.SNMP write, network configuration changes.
Supplemental gateway / controller evidenceOptional SSH operational read commands in the actual routing/controller context.Configuration backup account replacement, blanket administrator access.
Windows Security eventsDedicated domain user with Security-channel read rights, plus WinRM remote-shell execution rights on selected DCs.Domain Admin, Enterprise Admin, AD writes or permission to clear logs.
Appliance inventory accessudt.view.Access to user activity.
Usernames and login activityudt.view_users or udt.manage.Sharing user activity with every inventory viewer.
Configuration / port actionsudt.manage; direct-switch actions additionally require a deliberately provisioned write-capable SNMP credential.Write access for normal discovery; controller-child port actions.