Solved! Go to solution.
The thing is, it's not an error per se. The wireless client is roaming between APs so fast that the switch is essentially seeing the same MAC address out two or more switchports. I see the same at large warehouses, where clients, in this case scanners, are onboard those palletmovers, that can do almost 40 kilometers per hour. That's a quick roam. Afair, it's the MAC address table that gets a new entry from the same client on a new AP, before the old MAC address table entry ages out.
If you have Cisco Classic with APs in Flexconnect Mode, you'll see the exact same thing, and on Cisco Classic the "fix" here is to simply suppres the MAC flapping event.
But like I said, instead of bridging directly to a VLAN on the switchport (which in reality is the same concept as Flexconnect), configure Client IP Assignment to Layer 3 Roaming with a Concentrator, and concentrate traffic on the MX instead. This worked for me.
It may also work with Layer 3 Roaming only, but I haven't tried this yet.
One way I've found to mitigate those log messages, is changing the Client IP Assignment to Layer 3 with a Concentrator, and concentrate traffic to the MX.
I have yet to conclude if there is a performance boost, though.
It's purely cosmetic. It might be a good idea that you list your ports where there are access points and just ignore those messages from those ports.
The thing is, it's not an error per se. The wireless client is roaming between APs so fast that the switch is essentially seeing the same MAC address out two or more switchports. I see the same at large warehouses, where clients, in this case scanners, are onboard those palletmovers, that can do almost 40 kilometers per hour. That's a quick roam. Afair, it's the MAC address table that gets a new entry from the same client on a new AP, before the old MAC address table entry ages out.
If you have Cisco Classic with APs in Flexconnect Mode, you'll see the exact same thing, and on Cisco Classic the "fix" here is to simply suppres the MAC flapping event.
But like I said, instead of bridging directly to a VLAN on the switchport (which in reality is the same concept as Flexconnect), configure Client IP Assignment to Layer 3 Roaming with a Concentrator, and concentrate traffic on the MX instead. This worked for me.
It may also work with Layer 3 Roaming only, but I haven't tried this yet.
I can partially agree with you there. Maybe Meraki will manage to suppress flapping for wireless clients when an MR is connected to a port. With a WLAN bridge, the information is again very helpful ;=)
Then I will include it in our documentation for now.
Thank you so much 🙂