Connecting Extreme Network switches to HPE FlexFabric - Issues with MLAG possibly

Hi,

We have two extreme network core switch stacks. (made up of two switches each). They are connected together using load sharing or port channel to us normal people. And use MLAG to connect to a number of distribution switches, also extreme networking. There is no spanning tree due to EN stating not to use it when having MLAG in place.

We are connecting in a new HPE FlexFabric 5700 switch stack (made up of 2 switches). The ideal is to connect two ports to one core stack, and two ports to the other core stack.

We have setup the LACP connectivity on the HPE side, having two bridge aggregation ports. Connecting from two ports on the HPE to one core switch. Works fine. LACP etc all done. When you connect the other two ports to the second core switch, you get a network loop.

Now I believe the right way to do this, is to have on the HPE switch, one bridge aggregate made up of 4 ports. Connect two ports to each core switch stack. And configure the core switch stack ports as MLAG members.

Anyone any suggestions or am I in the right area?

EDIT: Rang ExtrememNetworkis, basically the answer is one Bridge Agg on the HPE with all 4 ports. Setup sharing on the extreme switch (2 ports per switch) with LACP. then mark the ports as MLAG

enable mlag port x:x peer "peer name" id X

and voila