LACP Failing to TP Link SG3424 Switch



  • I'm trying to set up a LACP LAG between pfsense and a TP-LINK SG3424, but the link just seems to come up briefly and then drop and continues to flap between these states.  Has anyone else tried this with these switches or had similar issues with others?

    I've sucessufully tested the configuration between the pfsense device and another manufactures switch (MiLan MIL-SM801P).  I've also been able to sucessfully create a LACP link between the Tp-Link and the MiLan switches.  I just can't get it to be stable between the TP-Link and the Pfsense device.  My gut feeling is that this is a bug with the TP-Link code, but it does seem to work fine with the other switch hence further investigation needed.

    Everything is stripped back to bare bones:

    • 1 interface on the LAG (em1)
    • No vlans on the LAG

    Plug em1 into the MiLan swtich - LAG comes up striaght away

    lagg0: flags=8843 <up,broadcast,running,simplex,multicast>metric 0 mtu 1500
            options=4219b <rxcsum,txcsum,vlan_mtu,vlan_hwtagging,vlan_hwcsum,tso4,wol_magic,vlan_hwtso>ether 00:1b:21:c2:85:5f
            inet6 fe80::21b:21ff:fec2:855f%lagg0 prefixlen 64 scopeid 0x8
            nd6 options=3 <performnud,accept_rtadv>media: Ethernet autoselect
            status: active
            laggproto lacp
            laggport: em1 flags=1c <active,collecting,distributing>Plug instead into the TP-Link and it flaps between:

    lagg0: flags=8843 <up,broadcast,running,simplex,multicast>metric 0 mtu 1500
            options=4219b <rxcsum,txcsum,vlan_mtu,vlan_hwtagging,vlan_hwcsum,tso4,wol_magic,vlan_hwtso>ether 00:1b:21:c2:85:5f
            inet6 fe80::21b:21ff:fec2:855f%lagg0 prefixlen 64 scopeid 0x8
            nd6 options=3 <performnud,accept_rtadv>media: Ethernet autoselect
            status: active
            laggproto lacp
            laggport: em1 flags=1c <active,collecting,distributing>and

    lagg0: flags=8843 <up,broadcast,running,simplex,multicast>metric 0 mtu 1500
            options=4219b <rxcsum,txcsum,vlan_mtu,vlan_hwtagging,vlan_hwcsum,tso4,wol_magic,vlan_hwtso>ether 00:1b:21:c2:85:5f
            inet6 fe80::21b:21ff:fec2:855f%lagg0 prefixlen 64 scopeid 0x8
            nd6 options=3 <performnud,accept_rtadv>media: Ethernet autoselect
            status: active
            laggproto lacp
            laggport: em1 flags=0<>

    Testing whether a LACP link to TP-Link works at all, from the TP-Link switch I see:

    When pfsense em1 connected TP-Link

    SWT-TEST#show lacp 5 neighbor
    Flags:  S - Device is requesting Slow LACPDUs
            F - Device is requesting Fast LACPDUs
            A - Device is in active mode      P - Device is in passive mode

    Channel group 5
                      LACP port                  Admin  Oper  Port    Port
    Port      Flags  Priority  Dev ID          Key    Key    Number  State

    Output flaps between
    Gi1/0/16  SA      32768      001b.21c2.855f  0      0x110  0x2    0xd
    &
    Gi1/0/16  SA      32768      001b.21c2.855f  0      0x110  0x2    0x3d

    When MiLan switch connected TP-Link

    SWT-TEST#show lacp 5 neighbor
    Flags:  S - Device is requesting Slow LACPDUs
            F - Device is requesting Fast LACPDUs
            A - Device is in active mode      P - Device is in passive mode

    Channel group 5
                      LACP port                  Admin  Oper  Port    Port
    Port      Flags  Priority  Dev ID          Key    Key    Number  State
    Initial state
    Gi1/0/16  SA      1          0040.c830.7b50  0      0x65  0x7    0xd
    Then stays stable on
    Gi1/0/16  SA      1          0040.c830.7b50  0      0x65  0x7    0x3d

    I don't belive the differing Priority, Operator Key and Port Numbers between these should be a factor.  So the next step for me is to try and capture the LACP traffic for each scenario to see if I can work out what is different between them and why this might happen.  Any advice on other things to try or gotchas I may have missed before I head down this path?

    N.B. I'm currently trying this on 2.1Beta but I've had the same issue on 2.0.1 as well.</performnud,accept_rtadv></rxcsum,txcsum,vlan_mtu,vlan_hwtagging,vlan_hwcsum,tso4,wol_magic,vlan_hwtso></up,broadcast,running,simplex,multicast></active,collecting,distributing></performnud,accept_rtadv></rxcsum,txcsum,vlan_mtu,vlan_hwtagging,vlan_hwcsum,tso4,wol_magic,vlan_hwtso></up,broadcast,running,simplex,multicast></active,collecting,distributing></performnud,accept_rtadv></rxcsum,txcsum,vlan_mtu,vlan_hwtagging,vlan_hwcsum,tso4,wol_magic,vlan_hwtso></up,broadcast,running,simplex,multicast>



  • FYI - This seems to have been an issue with the switch.  The recent firmware release (TL-SG3424_V1_130609) has a statement of  "Improved the LACP function of TL-SG3424."

    After upgrading the switch, it now works.  Finally I can deploy the config I've been wanting! :-)



  • Thats funny…    :P

    I guess "working" is an improvement.


Log in to reply