This applies to overlay logical switches only.

Procedure

  1. On the hypervisor, run the command /bin/nsxcli to start NSX CLI.
  2. Run the following command to see if the logical switch is present on the host.
    host1> get logical-switches
  3. Check that the state of the port is not admin down.

    On ESXi, run net-dvs and look at the response. For example,

    port 63eadf53-ff92-4a0e-9496-4200e99709ff:
    com.vmware.port.extraConfig.opaqueNetwork.id = … <- this should match the logical switch UUID
    com.vmware.port.opaque.network.id = …. <- this should match the logical switch UUID
    com.vmware.port.opaque.network.type = nsx.LogicalSwitch , propType = RUNTIME
    com.vmware.common.port.block = false, ... <- Make sure the value is false.
    com.vmware.vswitch.port.vxlan = …
    com.vmware.common.port.volatile.status = inUse ... <- make sure the value is inUse.

    If the logical port ends up in the blocked state, collect the technical support bundles and contact VMware support. In the meantime, run the following command to get the DVS name:

    [root@host1:~] net-dvs | grep nsx-switch
    com.vmware.common.alias = nsx-switch , propType = CONFIG

    Run the following command to unblock the port:

    [root@host1:~] net-dvs -s com.vmware.common.port.block=false <DVS-NAME> -p <logical-port-ID>

    On KVM, run ovs-vsctl list interface and verify that the interface with the corresponding VIF UUID is present and admin_state is up. You can see the VIF UUID in OVSDB in external-ids:iface-id.