Is it normal that the Sonos full mesh Spanning Tree Topology is always changing? We have about 20 - 30 Sonos devices communicating together and two Sonos boost devices that plug into our network.

Every 4 minutes a BPDU with the Topology Change notification is sent from either one of the Sonos boost devices causing the edge switch on the production network to flush its CAM table and forward the BPDU with the TCN bit set to the root bridge on our network which will also flush it’s CAM table. The switches on our network run 802.1w (RSTP) so when the BPDU with TCN flag is set the switches set their aging timer from 300 seconds to 15 seconds and flush out their CAM table on all ports except for the port that received the BPDU and also ports that are actively communicating.

I really don’t like the constant Spanning Tree Topology changes coming from the Sonos full mesh network affecting switches that are used on a enterprise network.  What is the best way around this? Would using 802.1s (MST) work around this so BPDU’s from production are not compatible with BPDU’s from Sonos STP? And then configure all ports to simply pass old STP BPDU’s so Sonos can still prevent loops with it’s proprietary implementation of STP?

If I completely unplug the Sonos Boost devices from our network, then the spanning tree topology stays static on our network and we can go well into 45 minutes+ and pretty much never get any Spanning Tree Topology changes.

What is the best practice when plugging Sonos into a production network?