Basic Lacp help

[Steven Webb]

I have a very basic setup
Server 1 Dell 710 with 4 port nic 6x 1Tb zfs
Server 2 same but has upgraded controllers and 50Tb storage 6x 10tb also using zfs not hardware raid.

Nic1 on bother servers is 192.168.1. 1 and 2 respecrivly..
I use this port for manegment of each

I then made a 3 port lacp or lagg on both. I tried back to back cables, best buy routers nothing worked.
Got a Cisco 4500 that now works I made 2 other channels 3 in each (one for each server) I then made a new.subnet 192.168.3. 1 and 2 tespectivly.

Switch now showes up and I can enter the admin interface and guide shows up I ssh in or use shell and I can ping across the lagg

This is when I simply setup the smaller server to create zfs snapshots which works then in setup to replicate across the link which does work
However I then loaded data on the server (3tb 300,000 files) the data is on 8 datasets. The total link at best is 3.6mbs which is slower than if I do the same setup but with 1 port that nets me 9.4mbs

I thought segmenting the relocation I would have a faster data movement as I'm asking it not to move faster just more.

Anyway is there something I can do to see if my Cisco is setup wrong (I just used defaults).or.am I missing something and this is correct?

 

[dlavigne]

I think you are seeing this:

LACP reads the sender and receiver IP addresses and, if they are deemed to belong to the same TCP connection, always sends the packet over the same interface to ensure that TCP does not need to reorder packets. This makes LACP ideal for load balancing many simultaneous TCP connections, but does nothing for increasing the speed over one TCP connection.