When troubleshooting a suspect port or verifying new hardware, a fiber-optic loopback test gives you a fast, definitive answer on whether an interface is healthy. The methodology is simple: start at the physical layer and work your way up the stack, confirming each layer before moving to the next. If a step fails, you’ve found your fault — no need to go further. Keep this guide and your test plug handy when uplinking switches, turning up a new dedicated internet circuit (that you ordered with Full Span Solutions), or lighting a short span of dark fiber.
Warning: Check Your Optic Type Before Inserting a Loopback Test Plug
⚠️ A direct fiber-optic loopback test plug is safe for standard SR and LR optics, but can permanently damage high-power transceivers. Before proceeding, confirm you understand what type of optic is installed in the port you intend to test.
The risk is straightforward: a loopback plug connects the Tx fiber directly to the Rx fiber with zero attenuation. If the transceiver’s maximum transmit power exceeds its own receiver overload threshold, the full laser output will be dumped directly into the receiver, potentially burning it out instantly.
Safe optic types for direct loopback use:
- 10GBASE-SR — Multimode, low power (~-7 to -1 dBm Tx). Safe for direct loopback.
- 10GBASE-LR — Single-mode, moderate power (~-8 to +0.5 dBm Tx). Safe for direct loopback. The IEEE 802.3ae standard defines the maximum Tx power and Rx overload threshold at the same value (+0.5 dBm).
- 1G SX / LX / ZX standard optics — Generally safe, but always verify the datasheet.
- 100GBASE-LR4 — Single-mode, LC duplex. Uses four wavelengths (LAN-WDM) over a single fiber pair. The IEEE 802.3ba standard sets the max Tx and Rx overload at the same level (+4.5 dBm), identical logic to 10GBASE-LR. Standard LC loopback plug.
- 100GBASE-CWDM4 — Single-mode, LC duplex, 2km reach. Similar power levels to LR4 and safe for direct loopback.
- 100GBASE-DR — Single-mode, LC duplex, 500m. Low power and safe for direct loopback.
Optic types that may require an attenuator or should NOT be looped directly:
- DWDM tunable optics — Can transmit at +3 dBm or higher, well above typical receiver overload thresholds.
- ZR / ZR+ / ER optics — Designed for extended reach and often transmit at high power levels (+1 dBm and above).
- Amplified or EDFA-backed optics — Output power can be extremely high and will almost certainly damage the receiver.
- Any optic with a Tx power spec above +1 dBm — Treat with caution and consult the datasheet before looping.
Confirming Your Optic Type
First, check to see what optics are installed in your device. You would pull the optic and look at the part number on its case. Alternatively, you could use the “show inventory” or equivalent command on your platform to see the optic that’s installed. If you are using an SR or another optic designed for multimode fiber, you will want to use the blue loopback plug. An LR or a single-mode optic will require the yellow plug.
Port Manufacturer Model Serial Number Rev ---- ---------------- ---------------- ---------------- ---- 1 XXXXXXX XXXXXXX SFP-10G-LR ABCW1234567890 0002
If the optic is already installed and the port is active, you can read the Tx power from the DOM data before inserting the fiber-optic loopback test plug:
Router#show interfaces ethernet 1 transceiver
Look at the Optical Tx Power (dBm) column. Then pull the datasheet for your specific optic and compare the maximum Tx power against the maximum Rx input (overload) power. If the maximum Tx power exceeds the maximum Rx input power, you must use an inline optical attenuator of sufficient value to bring the signal within the receiver’s safe operating range before using a loopback plug.
If you are unsure about the optic type or cannot locate the datasheet, do not use a bare loopback plug. Use an attenuating loopback (one with a built-in fixed attenuation value, typically 10–15 dB) as a safer default, or consult the manufacturer.
Confirming Your Fiber-Optic Loopback Test Plug
Most optics we see today use LC connectors for compact simplicity. However, there are still two types of fiber available, single-mode and multimode, depending on your application. Always consult your data sheet for specifics; however, generally, loopback plugs are typically yellow for single-mode or light blue for multimode. LR-type optics expect single-mode fiber, and SR-type optics expect multimode.
Step 1 — Confirm the Transceiver Is Receiving Light
The first question is the most fundamental: is the laser firing, and is the receiver seeing light? This is answered by the transceiver DOM (Digital Optical Monitoring) data, which is read directly from the SFP module’s internal monitoring circuits.
Router#show interfaces ethernet 1 transceiver
If device is externally calibrated, only calibrated values are printed.
N/A: not applicable, Tx: transmit, Rx: receive.
mA: milliamperes, dBm: decibels (milliwatts).
Bias Optical Optical
Temp Voltage Current Tx Power Rx Power
Port (Celsius) (Volts) (mA) (dBm) (dBm) Last Update
----- --------- -------- -------- -------- -------- -------------------
Et1 24.66 3.38 34.06 -4.53 -3.07 0:00:01 ago What a failure looks like: If the Rx Power column shows N/A, a very low value like -40dBm, or does not change after inserting the plug, either the loopback plug is not seated properly, the SFP’s laser is not firing, or the SFP’s receiver is damaged.
Step 2 — Confirm the Interface Is Up
With light confirmed at the physical layer, the next check is whether the interface has come up. A working loopback plug should bring the line protocol up on most platforms within a few seconds of insertion.
Router#show interface ethernet 1 status Port Name Status Vlan Duplex Speed Type Flags Encapsulation Et1 connected 1 full 10G 10GBASE-LR
The connected status confirms the port has established a valid link. The speed and duplex (10G, full) are correctly negotiated, and the type 10GBASE-LR confirms the device has the correctly identified the installed optic.
For more details, the full interface output confirms the same:
Router#show interface ethernet 1 Ethernet1 is up, line protocol is up (connected) Hardware is Ethernet, address is 0000.0000.0001 (bia 0000.0000.0001) Ethernet MTU 10178 bytes, Ethernet MRU 10200 bytes, BW 10000000 kbit Full-duplex, 10Gb/s, auto negotiation: off, uni-link: disabled Up 16 seconds Loopback Mode: None
The key line is:
Ethernet1 is up, line protocol is up (connected)
Note also that Loopback Mode : None confirms this is a physical loopback plug test and not a software-configured loopback — meaning the full transceiver and optical path are being exercised.
What a failure looks like: If the interface remains notconnect after inserting the plug, the SFP is likely faulty or not seated correctly. If the port shows errdisabled, check whether port security or a BPDU guard policy is interfering.
Step 3 — Confirm Counters Are Incrementing and Errors Are Zero
With the interface up, the next step is to verify that traffic is actually flowing through the loop cleanly. You want to see input and output counters incrementing together, and all error counters at zero.
Router#show interfaces ethernet 1 counters Port InOctets InUcastPkts InMcastPkts InBcastPkts Et1 42876 0 335 0 Port OutOctets OutUcastPkts OutMcastPkts OutBcastPkts Et1 42876 0 335 0
What a failure looks like: Rising CRC or FCS errors with the loopback in place indicate a problem with the transceiver itself, even if Rx power looks acceptable. Counters that increment on output but not input (or vice versa) suggest a faulty loopback plug or a failed receiver.
Step 4 — Confirm End-to-End Frame Delivery via LLDP
The final and most conclusive test is to confirm that complete Ethernet frames are being transmitted, looped, and successfully processed by the switch — all the way up the stack. LLDP (Link Layer Discovery Protocol) provides a perfect passive test for this, since it runs automatically on most switches without any configuration.
Enabling LLDP on Most Platforms
LLDP is enabled globally by default on many switches & routers, but if it has been disabled or if LLDP has been turned off on a specific interface, you can re-enable it with the following commands:
Enable LLDP globally:
Router(config)#lldp run
Enable LLDP on a specific interface (if it was disabled at the interface level):
Router(config)#interface ethernet 1 Router(config-if-Et1)#lldp transmit Router(config-if-Et1)#lldp receive
Verify LLDP is running:
Router#show lldp
The output will show “LLDP state: Enabled” if LLDP is active globally. Allow up to 30 seconds after enabling LLDP for a neighbor entry to appear, as that is the default LLDP transmit interval. If you need faster results, you can clear the LLDP table and trigger an immediate retransmit with
clear lldp neighbors
Once LLDP is confirmed running, you can check the LLDP neighbors table. The device should see itself.
Router#show lldp neighbors Last table change time : 0:00:21 ago Number of table inserts : 3 Number of table deletes : 2 Number of table drops : 0 Number of table age-outs : 0 Port Neighbor Device ID Neighbor Port ID TTL ---------- ------------------------ ---------------------- --- Et1 Router Ethernet1 120
What a failure looks like: If the LLDP neighbor table shows nothing on Ethernet1 despite the interface being up and counters incrementing, the port may have LLDP disabled, or frames are being corrupted in transit (which would show up as CRC errors in Step 3).
Summary
| Step | Command | What You Are Verifying |
|---|---|---|
| 1 — Physical layer | show interfaces ethernet 1 transceiver | The laser is firing, the Rx side sees light |
| 2 — Link layer | show interface ethernet 1 status | Interface is up and connected |
| 3 — Frame integrity | show interfaces ethernet 1 counters | Traffic flowing, zero errors |
| 4 — Data plane | show lldp neighbors | Full end-to-end frame delivery confirmed |
A port that passes all four steps is healthy and ready for production. A port that fails at any step has a problem at that layer — and you don’t need to look any further up the stack to identify where the fault lies.