TL;DR: Test used network equipment in four steps: inspect and boot the unit, verify identity with show commands, loop-test every port, and burn in for 24-48 hours under load. If the seller already runs a documented testing process, most of this work is done before the unit ships.
Before powering anything on, check the chassis for damage, verify the serial number against the label, and confirm accessories (rack ears, power cables, console cable). A unit that was pulled from production with care looks it — heavy scratches and bent brackets often signal rough handling or repeated failures.
Connect power and a console cable, and watch the boot sequence. The unit should reach the login prompt without errors. Then verify identity and health:
show version — model, serial, software version, uptime.show inventory — installed modules and their serials.show environment / show chassis — temperatures, fans, PSU status.Cross-check the serials against the seller’s documentation and, for Cisco gear, against Cisco’s serial lookup tools to confirm the unit is genuine.
Every port must pass a traffic test. Use a loopback plug, or better, connect ports to a second switch and verify link up and traffic flow. For PoE models, confirm each PoE port powers a real device and the budget behaves under load. Skipping this step is how "one bad port" surprises arrive after deployment. The same test covers stack members like the WS-C3850-24P-E and fixed access switches alike.
Run the unit 24-48 hours with all ports active and traffic passing. This is the phase where weak power supplies, failing fans and marginal memory are exposed. The seller’s quality control process should include exactly this: boot, console, per-port tests, PoE budget and burn-in, with the report available on request.
Looking for reliable used network equipment at a fair price? Send us the model list and we will reply with availability and pricing within 24 hours.
| Section | What it confirms |
|---|---|
| Unit identity | Serial number matches label and invoice |
| Boot results | Clean boot to console, no error logs |
| Port tests | Every port passed loopback, link and traffic test |
| PoE budget | All PoE ports deliver rated power |
| Burn-in | Hours under load with stable temperature and fans |
| Software | Installed IOS/Junos/VRP version and upgrade path |
A: Connect a loopback plug (or simply patch two ports together with a short cable) and check the interface state in the CLI. For a thorough test, connect each port to a tester or another switch and generate traffic. A documented report from the seller covers all ports for you.
A: Industry practice is 24 to 48 hours under load with all ports active. This catches intermittent PSU and fan failures and weak memory modules that only appear after warm-up. Serious dealers run this burn-in before shipping.
A: Not strictly — a unit that boots to the login prompt and serves DHCP is a good sign — but a console session lets you read show version, show inventory and the full port state, which makes verification far more reliable. Console cables are cheap and included with most tested units.
A: Power a PoE device (phone, camera or AP) on each PoE port and confirm it draws power and the budget is consumed correctly in show power inline. Also verify the PoE budget under multiple devices, since failing power supplies often limit the actual budget.
A: The unit serial, power-on and console boot results, every port’s loopback result, PoE budget checks, burn-in hours, cosmetic grade and the installed software version. If a seller cannot produce this, treat the unit as untested.
A: A reputable dealer covers DOA with a replacement or refund under the warranty. Document the issue with photos and a console log within the stated DOA window (typically 7-14 days) and the claim is straightforward.
Need help deciding which pre-owned model fits your network? Contact our engineers or message us directly on WhatsApp — we answer technical questions for free.