Interface Down? Work the Problem in the Right Order
A layer-by-layer checklist for troubleshooting physical and logical interfaces that will not come up.
Read the interface state literally
Administrative state, operational state, last-flap time, and protocol state tell you where to start. An administratively down port is a different problem from a port that is enabled but has no carrier.
Do not stop at the summary line. Check transceiver presence, negotiated speed, duplex, MTU, error counters, alarms, and whether the remote device sees the same condition.
Prove the physical layer before blaming routing
For copper, verify patching, speed, duplex, and the far-end port. For fiber, compare local and remote transmit/receive levels, optic type, wavelength, reach, connector cleanliness, and polarity. Light being present does not automatically mean it is within the receiver operating range.
- Confirm local Tx/Rx.
- Confirm far-end Tx/Rx.
- Verify optic and wavelength compatibility.
- Check patching and polarity before replacing hardware.
Compare configuration and substitute one component at a time
A healthy physical link can still fail because the endpoints disagree on speed, encapsulation, VLAN mode, LACP, MTU, or shutdown state. A focused config diff is faster than reading two full configurations independently.
If substitution is needed, swap one known-good component at a time and preserve the original. If the fault follows the optic or jumper, you have evidence; if it stays with the port or fiber path, continue narrowing there.