/var/log/syslog
[TIMESTAMP] <Edge-Node-Name> NSX 3631 SYSTEM [nsx@6876 comp="nsx-edge" subcomp="frr-config" username="frr" level="ERROR"] "Exception while generating FRR Config: ['Traceback (most recent call last):\n', ' File "/opt/vmware/nsx-edge/bin/frr.py", line 536, in apply_config_to_frr\n if generate_frrcfg_txt_from_pb(routing_cfg_msg):\n', ' File "/opt/vmware/nsx-edge/bin/frr.py", line 406, in generate_frrcfg_txt_from_pb\n pim.mcast_cfg(frr_cfg)\n', ' File "/opt/vmware/nsx-edge/python/com/vmware/nsx/edge/frr/pim.py", line 121, in mcast_cfg\n print_mcast_interfaces(frr_cfg, mcast_config)\n', ' File "/opt/vmware/nsx-edge/python/com/vmware/nsx/edge/frr/pim.py", line 66, in print_mcast_interfaces\n if ((mcast.mcast_role == rcpb.MulticastConfigMsg.MCAST_ACTIVE and \\\n', "UnboundLocalError: local variable 'mcast' referenced before assignment\n"]"
[TIMESTAMP] <Edge-Node-Name> NSX 3631 SYSTEM [nsx@6876 comp="nsx-edge" subcomp="frr-config" username="frr" level="ERROR"] "Error in applying the config to FRR"
VMware NSX-T Data Center 3.x
VMware NSX 4.x
After disabling Multicast on a T0, the Multicast HA role still remains as Active on the Active Edge and Standby on the Standby Edge. On the Active Edge, RCPM continues to push multicast configuration to the Edge which causes the datapath outage.
Fixed in NSX-T Data Center 3.2.4
Fixed in NSX 4.2
Workaround:
Reboot or place Edges in/out of Maintenance Mode. This will clear the Multicast configuration from the Edges. The issue will not reoccur provided Multicast is not enabled/disabled again.