After node reboots, LACP ports are down and switch ports are in 'Link Flap error-disable' state

Applies to

  • ONTAP
  • Interface group using Link Aggregation Control Protocol (LACP)

Issue

  • Storage nodes are rebooted during ONTAP upgrade and some LACP members are in a down state.

::> ifgrp show -fields down-ports,up-ports

  • A node's network ports are down and in a no link state.

::> network port show -node <node_name>

  • ​​​​​​​There is no indication from storage side of any link flapping or any attempt to negotiate LACP measured from Active IQ IFCONFIG-VVV section or node run -node <node> ifconfig -v:

a0a: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 9000
lagg statistics:
active ports: 0
flapping: 0
laggport: e0a flags=20000<DEFAULT> state=41<ACTIVITY,DEFAULTED>
        input/output LACPDUs: 0 / 0
laggport: e2c flags=20000<DEFAULT> state=41<ACTIVITY,DEFAULTED>
        input/output LACPDUs: 0 / 0
status: no carrier

  • The Cisco network switch is investigated and it is determined that the corresponding switch interfaces are in the Link Flap error-disable state.
  • Cisco switch logging shows repeated link failures and disables:

%ETHPORT-2-IF_DOWN_LINK_FAILURE: Interface Ethernet1/1/2 is down (Link failure)
%ETHPORT-2-IF_DOWN_LINK_FAILURE: Interface Ethernet1/1/2 is down (Link failure)
%ETH_PORT_CHANNEL-8-PORT_SUSPENDED: Ethernet1/1/2: Ethernet1/1/2 is suspended.
%ETHPORT-2-IF_DOWN_LINK_FAILURE: Interface Ethernet1/1/2 is down (Link failure)
%ETHPORT-2-IF_DOWN_LINK_FAILURE: Interface Ethernet1/1/2 is down (Link failure)
%ETHPORT-2-IF_DOWN_ERROR_DISABLED: Interface Ethernet1/1/2 is down (Error disabled. Reason:Too many link flaps)