Host Flapping - Brought entire campus network down

[William Reed]

Hello,

Yesterday we had a desktop tech patch a linksys switch into 2 different access ports on 2 different 3560s on our campus. This in turn created a major problem in which the entire network was acting sporadic. The core 4510 CPU spiked to 99% and sat there, also all the trunk ports were transmitting 90mpbs consistently throughout the campus.

I have BPDUGUARD enabled on all the switches by default.

My question is why didn't the Cisco switches sense this and shut one of the ports off? The second I shut the culprit port down, the network went back to normal.

How do I prevent this in the future? It took 2 hours to figure this out, in the mean time the network was basically down. I need to know how to prevent this in the  future.

[Leo Laohoo]

I have BPDUGUARD enabled on all the switches by default.

Check the configuration of the port.  Maybe BPDU Guard is enabled on an access port but the Linksys switch could be a Trunk port so BPDU Guard wouldn't take any effect on this.

[William Reed]

Leo,

Both ports he plugged into were set to:

'switchport mode access'     and

'spanning-tree portfast'

'spanning-tree portfast bpduguard default' are enabled on all the switches.

He was not plugged into a Trunk port.

[Madhukrishnan Gopinathan Nair]

William,

Do you have bpdu filter enabled on these ports?

Thanks,

Madhu

[William Reed]

Anyone else?

[William Reed]

Anyone else?

[Madhukrishnan Gopinathan Nair]

William,

Just posting the below post again since i do not see an aswer.

Do you have bpdu filter enabled on these ports?

Thanks,

Madhu

[William Reed]

Anyone else? I dont think replies are being posted to the thread.