LACP lagg not distributing traffic

[Razorblade]

Hello,

I finally managed to get a LACP lagg working with my FreeNAS 8.3.1 box.Two members, re0 and em0.

Failover works fine - when I pull out the cable from port one, the second port will take over the connection while only losing a few packets.

But, when I try to distribute traffic among the two lagg members this won't work.
I tried to download a file via FTP from the FreeNAS box on client 1 and at the same time upload a file via SMB to the FreeNAS box from client 2. The traffic will only pass through the first NIC of the lagg - the LED of the second NIC does not indicate any traffic.

ifconfig looks okay:

Code:

re0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
    options=209b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,WOL_MAGIC>
    ether c8:60:00:58:4a:6d
    media: Ethernet autoselect (1000baseT <full-duplex>)
    status: active
em0: 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 c8:60:00:58:4a:6d
    media: Ethernet autoselect (1000baseT <full-duplex>)
    status: active
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384
    options=3<RXCSUM,TXCSUM>
    inet6 fe80::1%lo0 prefixlen 64 scopeid 0xa 
    inet6 ::1 prefixlen 128 
    inet 127.0.0.1 netmask 0xff000000 
    nd6 options=3<PERFORMNUD,ACCEPT_RTADV>
lagg0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
    options=209b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,WOL_MAGIC>
    ether c8:60:00:58:4a:6d
    inet 192.168.12.3 netmask 0xffffff00 broadcast 192.168.12.255
    media: Ethernet autoselect
    status: active
    laggproto lacp
    laggport: em0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
    laggport: re0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>

But the daily reports show some errors on the second interface:

Code:

Network interface status:
Name    Mtu Network       Address              Ipkts Ierrs Idrop    Opkts Oerrs  Coll
re0    1500 <Link#1>      c8:60:00:58:4a:6d   426559     0     0  1508707     0     0
em0    1500 <Link#2>      c8:60:00:58:4a:6d  4170985     0     0 14769041     0     0
usbus     0 <Link#3>                               0     0     0        0     0     0
usbus     0 <Link#4>                               0     0     0        0     0     0
usbus     0 <Link#5>                               0     0     0        0     0     0
usbus     0 <Link#6>                               0     0     0        0     0     0
usbus     0 <Link#7>                               0     0     0        0     0     0
usbus     0 <Link#8>                               0     0     0        0     0     0
usbus     0 <Link#9>                               0     0     0        0     0     0
lo0   16384 <Link#10>                           2648     0     0     2648     0     0
lo0   16384 fe80::1%lo0   fe80::1                  0     -     -        0     -     -
lo0   16384 localhost     ::1                      6     -     -        6     -     -
lo0   16384 your-net      localhost             2642     -     -     2642     -     -
lagg0  1500 <Link#11>     c8:60:00:58:4a:6d  4554517     0     0 16232887    34     0
lagg0  1500 192.168.12.0  192.168.12.3       4496819     -     - 16230497     -     -

Those 34 Oerrs appear every day.

Both NICs have the same priority.

How can traffic be distributed among all of the lagg members?

 

[Setius]

LACP does not provide per packet load balancing, so you will never see even traffic flow on both nic's. The LB algorithm is done based on some type of source/destination which could be Mac address/IP address/subnet. Google can find you a more in depth answer if you want it, but wikipedia has a decent explanation.

http://en.wikipedia.org/wiki/Link_aggregation

 

[cyberjock]

...and now you know a small piece of the story as to why I have repeated said in the forum that unless you have lots of users LACP is useless. :)

To use the "extra" bandwidth that LACP offers you really DO need lots of users. Home users aren't going to see a benefit. I learned this lesson the hard way years ago.

Hopefully you didn't spend too much time/money trying to get LACP to work. :P