The aim is to setup LACP.


alias bond0 bonding
options bond0 miimon=100 mode=4 lacp_rate=1






switch#show etherchannel X summary
Flags:  D - down        P - bundled in port-channel
        I - stand-alone s - suspended
        H - Hot-standby (LACP only)
        R - Layer3      S - Layer2
        U - in use      f - failed to allocate aggregator

        M - not in use, minimum links not met
        u - unsuitable for bundling
        w - waiting to be aggregated
        d - default port

Number of channel-groups in use: 3
Number of aggregators:           3

Group  Port-channel  Protocol    Ports
X     PoX(SD)        LACP        Gi0/A(I)   Gi0/B(I)   Gi0/C(I)

interface Port-channelX
 switchport mode access


interface GigabitEthernet0/A
 switchport access vlan X
 switchport mode access
 channel-protocol lacp
 channel-group X mode active


Mar  2 11:17:50.529: %SW_MATM-4-MACFLAP_NOTIF: Host <MAC_ADDRESS> in vlan X is flapping
between port Gi0/Y and port Gi0/Z


Once the network service has been restarted on the Linux server the slave ethernet adapters are up and the system is still accessible:

  1. Why does the Cisco log indicate that the host is flapping between the ports?
  2. Why could the system be accessed while the portchannel is down?
  • The (I) in the show etherchannel summary means exactly what it says in the key: I - stand-alone. Since the ports are acting as stand-alone interfaces, they are not bundled into the etherchannel and the mac-flapping is expected. You need to fix the channel configuration for this to work correctly, there are probably other log messages that provide more detail for this. Also the output of show etherchannel ## detail should be helpful.
    – cpt_fink
    Mar 5, 2015 at 5:39
  • Did any answer help you? if so, you should accept the answer so that the question doesn't keep popping up forever, looking for an answer. Alternatively, you could provide and accept your own answer.
    – Ron Maupin
    Aug 10, 2017 at 4:30

2 Answers 2


If the port channel is DOWN, then the Cisco device doesn't see it bundled together. As your display shows, each port are Stand-Alone (as indicated by the "I" status), meaning they're seen as separate entities to the switch with regards to MAC table, etc.

This point to something wrong in your channel config and/or capabilities (mismatch between the linux box and the switch)

I've often seen this with vMware servers that has some problems with their network config, the VMs are often balanced between their link using the DST IP, the switch seeing the VM's MAC on one interface, then another, and another, etc.

On the Cisco side, you can try to look-up some debug to help you ("debug etherchannel ?")

  • There is no VM or VMware. The system is a physical server.
    – 030
    Mar 2, 2015 at 14:28
  • There is a MAC sent by your linux box that is going on more than one link, either by bridging, switching, or using a virtual MAC. I've never touched bonding on Linux, but I know that Etherchannel between Cisco and something else can be problematic sometimes, depending on the actual setup. Mar 2, 2015 at 17:41

Your etherchannel is setup incorrectly. If you look in the logs, you'll see a configuration mismatch because the PoX and G0/X interfaces do not have identical functional configurations.

Start over. Delete PoX. Default the G0/X interfaces. The ONLY thing you configure on the member interfaces is the channel-group. Then setup everything on Port-channelX; those elements will be copied to all member ports.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.