Up until last evening, I had a (4) port LAG between 2xN4032s & 2x6224's each in an indpendant stack configuration. These switches are being used as a path from our VMWARE system to our corporate LAN and up until a reset of the N4032's last evening, the configuration was working well, atleast I thought.
I am leaning towards this being a Spanning Tree issue on the channel being utilized. When doing a show interface port-channel, the port-channel had been in a discarding state since last evenings reload of the stack. I'm looking for recommendations as to how I would go about ensuring the 4 ports being utilized in the path are configured correctly. As of this writing, i've backed out the LAG simply relying on a single port configured to pass the LAN VLAN traffic and it is working.
I had ports unit1 - 1/2 & unit2 - 1/2 of the N4032 connected to unit 1-5/6 & unit2 - 5/6 respectively. I had the port-channel mode in GENERAL adding the VLAN as a tagged member. I had portfast enabled as well as RSTP enabled. Any information as to how this should look between the 4 ports would be greatly appreciated.
The overall VM-LAN traffic path is as follows; HOST1 > STACKED-N4032 <(4PORT-LAG)> STACKED-6224 <(2PORT-LAG)> 6224 EDGE > Corporate LAN