We have been having an issue with LACP between our Aruba 6000 controller and our Juniper EX 8208 core switch. Our configuration includes Qty4 1Gb ports in an AE on the Juniper side, set to active and periodic fast with a link speed set to 1G, and Qty4 1Gb ports in a port-channel on the Aruba side set to active. Chronically, we will begin to receive LACP errors on one or more interfaces on the Juniper side saying the the interface state has changed (KERN_LACP_INTF_STATE_CHANGE).
Aug 28 09:22:19 C-CORE-XRE0 lacpd[1390]: LACPD_TIMEOUT: ge-1/0/46: lacp current while timer expired current Receive State: CURRENT
Aug 28 09:22:19 C-CORE-XRE0 member1-fpc3 PFE Manager received unknown trigger (type Queue, id 5)
Aug 28 09:22:19 C-CORE-XRE0 /kernel: KERN_LACP_INTF_STATE_CHANGE: lacp_update_state_userspace: cifd ge-1/0/46 - CD state - ready to carry traffic
Aug 28 09:22:19 C-CORE-XRE0 member1-fpc1 VCP TRUNK DEL: Trunk [25] -> (12/1)
Aug 28 09:22:19 C-CORE-XRE0 member1-fpc1 VCP TRUNK DEL: Trunk [25] -> (8/1)
Aug 28 09:22:19 C-CORE-XRE0 member1-fpc1 VCP TRUNK DEL: Trunk [25] -> (12/1)
Aug 28 09:22:19 C-CORE-XRE0 member1-fpc1 VCP TRUNK DEL: Trunk [25] -> (8/1)
Aug 28 09:22:19 C-CORE-XRE0 member1-fpc1 PFE Manager received unknown trigger (type Queue, id 5)
There are no errors to speak of on the Juniper side in the LACP counters:
joeuser@C-CORE-XRE0> show lacp statistics interfaces ae9Aggregated interface: ae9
LACP Statistics: LACP Rx LACP Tx Unknown Rx Illegal Rx
ge-17/0/45 3986783 504722 0 0
ge-1/0/45 3986782 504719 0 0
ge-17/0/46 3895407 710840 0 0
ge-1/0/46 3910126 677602 0 0
Or the Aruba side:
(C-ARUBA-A6000-01) # show lacp 0 counters
LACP Counter Table
Port LACPDUTx LACPDURx MrkrTx MrkrRx MrkrRspTx MrkrRspRx ErrPktRx
GE 0/0 22678448 12150922 0 0 0 0 0
GE 0/1 22601860 12323721 0 0 0 0 0
GE 0/2 22659397 12180711 0 0 0 0 0
GE 0/3 22586979 12353276 0 0 0 0 0
Besides causing some CPU overhead and filling up our logs, I do not think there is any adverse affect, but it is irritating. We are running ArubaOS 6.3.1.7 on the controller and 12.3R9.4 on the core switch. Has anyone experienced this issue or seen this problem before?