asked on
Cisco Bandwidth Aggregation
I am having an issue getting bandwidth aggregation to work properly between a Windows Server 2008, Cisco SG500-52, and Windows Server 2012 R2.
On the 2008 server I have two 1Gb ports configured with BACS4 with a Team Type of 802.3 ad. For these two ports the switch has them configured as trunk ports. The ports are also in one LAG, LACP is Enabled, the link is up, and it has these two ports as active members.
On the 2012 R2 server I have two 1Gb ports configured through Windows NIC Teaming. The Teaming mode is LACP and the load balancing mode is Dynamic. The status says that it is Online and it shows both ports as being Active. For these two ports the switch has them configured as trunk ports. The ports are also in one LAG, LACP is Enabled, the link is up, and it has these two ports as active members.
This configuration still limits my bandwidth to 1Gb. On both servers I have used Performance Monitor and looked at each individual NIC for Bytes Total/sec and it looks like it is still only using one NIC.
What am I missing?
Thanks.
On the 2008 server I have two 1Gb ports configured with BACS4 with a Team Type of 802.3 ad. For these two ports the switch has them configured as trunk ports. The ports are also in one LAG, LACP is Enabled, the link is up, and it has these two ports as active members.
On the 2012 R2 server I have two 1Gb ports configured through Windows NIC Teaming. The Teaming mode is LACP and the load balancing mode is Dynamic. The status says that it is Online and it shows both ports as being Active. For these two ports the switch has them configured as trunk ports. The ports are also in one LAG, LACP is Enabled, the link is up, and it has these two ports as active members.
This configuration still limits my bandwidth to 1Gb. On both servers I have used Performance Monitor and looked at each individual NIC for Bytes Total/sec and it looks like it is still only using one NIC.
What am I missing?
Thanks.
Network ManagementMicrosoft HardwareSwitches / Hubs
Last Comment
ollybuba
SOLUTION
THIS SOLUTION ONLY AVAILABLE TO MEMBERS.
View this solution by signing up for a free trial.
Members can start a 7-Day free trial and enjoy unlimited access to the platform.
ASKER CERTIFIED SOLUTION
THIS SOLUTION ONLY AVAILABLE TO MEMBERS.
View this solution by signing up for a free trial.
Members can start a 7-Day free trial and enjoy unlimited access to the platform.
*meant to say, 'windows side of things I have my team set to as SLA'
ASKER
If I change my teaming mode to static teaming vs LACP do o need to enable LACP on the switch? How about configuring an LAG?
At one time I had an slm2014 which is similar to the switch you are using and at that time I had configured a LAG for the SLA. So I can't speak directly to whether or not use LACP or LAG on the switch side but I am confident a LAG will work for you on the switch side with SLA on the server side.
ASKER
What I have found is that when you have LACP as the teaming mode in Windows Server 2012 R2, you also have to have LACP enabled on the LAG with Cisco. Makes sense. When you use the static teaming mode, you still have to setup a LAG but you don't enable LACP. You also can use NIC teaming between two servers. The catch with NIC teaming is that each interface acts in what I would consider half-duplex mode. At least with two interfaces, one port sends and one receives. Having traffic moving in both directions is the only way to achieve full bandwidth utilization of the team from what I've seen. I'm guessing that changes when you start adding even more interfaces.