Interface Flapping Due to Duplex Mismatch

I was troubleshooting a branch router that kept dropping its uplink connection every few minutes. Users complained about unstable VPN sessions and slow internet. At first glance, everything looked fine—no obvious errors in the config, and the interface was up. But the logs told a different story: interface flapping.

🔍 Diagnosis

I ran:

• show interfaces GigabitEthernet0/0 and noticed frequent line protocol down/up messages.
• show logging revealed recurring %LINK-3-UPDOWN events.
• show controllers hinted at CRC errors and late collisions.

That’s when I suspected a duplex mismatch between the router and the connected switch.

✅ Solution

Here’s how I resolved it:

1. Checked switch port settings: On the switch side, the port was set to auto-negotiation.
2. Forced duplex and speed on the router:

interface GigabitEthernet0/0 speed 100 duplex full

3. Matched settings on the switch: I manually set the switch port to speed 100 and duplex full to ensure consistency.
4. Monitored stability: After the change, I monitored the interface with:

show interfaces GigabitEthernet0/0 | include line protocol

No more flapping. CRC errors dropped to zero, and users reported stable connectivity.

📚 Reference That Helped

The Cisco Enterprise Troubleshooting Guide was a lifesaver. It walked through how duplex mismatches can cause intermittent connectivity, especially when one side is set to auto and the other is hardcoded.

This issue is sneaky because it doesn’t always show up as a hard failure—it’s more like a slow bleed that degrades performance over time. It’s a good reminder to always verify both ends of a link when troubleshooting flapping or CRC errors.

Want me to help you turn these into a troubleshooting guide or training post for your blog?