We have a fresh link made on RB333 with 3.0RC13 which should serve as a redundant link to another (made on 2.9.x MT). There are cisco switches on both sides and we use RTSP all worked fine until we replaced one old redundant link with the RB333 one. After the new link was connected a nice packet storm occured. It looks like the cisco switches didn’t receive RSTP PDUs through the new link so they didn’t break the bridge loop. I enabled RSTP on the MTs and it works now - but I would prefer to have the link fully transparent and leave the ciscos to decide which link to stop (less amount of devices to care about when configuring the STP and I am not sure I can easily detect the link state via SNMP on MT)
The setting is simple - EoIP tunnel made on the wlan interfaces and ethernet bridged with the tunnel. The same settings we use on 2.9.x in many times without problems with RSTP. Is this a known problem (a bug in ROS).