Biometric Machine Not Working? Troubleshooting Guide for Attendance Devices

A biometric attendance machine that stops working usually fails in one of a handful of predictable ways. Most of these are fixable without a service call — a few genuinely are not. This guide walks through the common faults in the order worth checking them, and flags where the problem stops being about one device and starts being about the whole fleet.

Device unresponsive

Start with power — a loose connector or a tripped socket accounts for more "dead" devices than any hardware fault. Confirm the network link if the device relies on one for real-time sync. If both check out, a restart resolves a surprising share of unresponsive devices, particularly ones that have been running continuously for weeks without a reboot.

Fingerprint not detected

Check the sensor surface first — dust, grease or residue on the scanning glass is the most common and the easiest to fix, and should be part of routine cleaning rather than a fault-response step. If the sensor is clean and detection still fails for a specific worker, the issue is usually enrolment quality or the underlying condition described in our comparison of fingerprint and face recognition attendance: worn, dry or damaged ridges from manual work. Re-enrolment with a better-quality capture resolves most of these; workers whose ridges are too worn for reliable fingerprint capture are candidates for a face-recognition device instead.

Face not recognised

Check lighting and angle before assuming an enrolment problem — direct backlighting or a steep camera angle causes more face-recognition failures than any single other factor. If positioning is correct and the same worker still fails to authenticate, the enrolment image itself is usually the cause: a single low-quality enrolment photo, or one captured months before a significant change in the worker's appearance, produces unreliable matching. Re-enrolling with two or three images under normal working-light conditions typically fixes this.

Device offline

Confirm the device still has a valid IP address and that nothing has changed on the network side — a DHCP lease expiry or a firewall rule update are common, quiet causes. Corporate firewalls blocking the ports a device needs for sync are a frequent culprit, particularly after a network security review that nobody thought to check against the attendance device list.

Time drift

A device with an incorrect internal clock will log attendance events with the wrong timestamp, which quietly corrupts shift and overtime calculations without any visible error. Most devices sync time automatically from the network; if drift keeps recurring, it usually points to a device with an unreliable internal clock or a network time source that is itself incorrect.

Data not syncing to software

If the device is capturing attendance locally but the software shows nothing, check the sync connection first — this is the same network and firewall check as an offline device, but the device may still appear "online" in a basic sense while the specific sync channel is blocked. Where the device supports offline buffering, confirm the buffer is not full; a device that has been offline for an extended period can accumulate more events than its buffer holds, silently dropping the oldest ones.

When it is one device versus a fleet problem

A single device with a single fault is a maintenance ticket. The pattern worth watching for is the same fault appearing across multiple devices at similar times — that usually points to a shared cause: a network change, a firmware update, a certificate expiry, or an environmental factor like a seasonal lighting change affecting every outdoor face terminal at once.

At that point, troubleshooting device by device is the wrong response. What is needed is fleet-level visibility — which devices are online, which have pending exceptions, which share the same firmware version — so the shared cause can be identified and fixed once rather than chased terminal by terminal.

When to call an engineer

Power, network, cleaning and re-enrolment cover the majority of faults and can be handled on-site without specialist support. A hardware fault inside the sensor or camera module, a device that won't hold a firmware update, or a fault that recurs immediately after a fix are the signals that it's a hardware or engineering issue rather than a configuration one.