Asset Sync and Duplicate Machines Why the Same PC Appears Twice

ManageEngine documents the flow plainly. Immediate: inventory details of computers and mobile devices, tickets created from the agent icon, deployment status updates, and templates. Every few minutes: approval requests for self-service and prohibited software — about every 3 minutes. Every 30 minutes: the currently logged-on user data. Once every 24 hours, or on demand: newly created software packages.

Memorise the last two lines and you will save yourself real panic. The user shown on an asset can legitimately be half an hour behind the person actually sitting there, and a software package built this morning may not be deployable from a ticket until the daily sync — unless you trigger it on demand.

ServiceDesk Plus can only display what Endpoint Central collected. Endpoint Central fetches inventory at system start-up, at user logon, after manual or scheduled scans, and whenever software is installed or uninstalled — and a rename or domain change triggers a fresh scan too.

So a stale record has two possible homes: the sync between products, or the scan on the machine itself. A PC that has been switched off in a storeroom for a month shows month-old data in both consoles, correctly. Before blaming the bridge, check whether the agent on the machine is alive and scanning.

Now the real problem. The same physical PC shows up twice in ServiceDesk Plus. The usual causes, in order of how often they are guilty:

1. The machine was renamed. The integration matches strongly on computer identity and name. A rename usually triggers a rescan and an update — but a machine renamed while offline, or reimaged and rejoined, can leave the old record standing next to the new one.
2. The machine existed in ServiceDesk Plus before integration. Assets entered manually or imported earlier do not always marry cleanly with the Endpoint Central record arriving later, especially if names differ even slightly.
3. The agent was reinstalled or the OS rebuilt. To the systems, the rebuilt machine can look like a new arrival while the old ghost record remains.

The fix is housekeeping, not hope: decide which record carries the true history, merge or retire the duplicate, and check the machine's name and agent health so it does not split again next month.

Most duplicate pain is naming pain wearing a costume. A fleet where computers are named by a fixed pattern — site, department, number — syncs cleanly for years. A fleet with names like PC, New PC, Ravi's Laptop Final generates a steady supply of almost-matching records, and every rename is a small integration incident.

If you are integrating an older, messier fleet, treat the first sync as an audit: export the asset list a week later, sort by name, and spend one focused hour merging the obvious pairs. It is unglamorous work that pays rent every single week afterwards, because technicians trust an asset list they have seen be right.

Renames are the integration's known stress point, so make them a procedure instead of a surprise: rename the machine while it is online, let Endpoint Central rescan (the rename itself triggers one), confirm the new name in Endpoint Central first, then confirm it in ServiceDesk Plus. If the old name lingers as a separate asset after a day, retire it deliberately rather than waiting for it to confuse a technician during a real ticket.