LACP does not provide more throughput but does provide more bandwidth. 
Basically if you have two 1G NICs in an LACP bond, you will not get a stream 
faster than 1G.. However, since you are not load balancing, you can have MORE 
1Gb streams.

So what your seeing is normal.
 
Also do a ovs-vsctl list port and a xe bond-list params=all to make sure your 
bonds are negotiating properly.


Regards,
Marty Godsey

-----Original Message-----
From: Alessandro Caviglione [mailto:[email protected]] 
Sent: Monday, August 22, 2016 4:02 PM
To: [email protected]
Subject: XenServer bond question

Hi guys,
pheraps I should post this question on Citrix forum but I think that here I can 
find a more "practical" answer based on similar infrastructure. :) I've a CS 
4.6 with XS 6.2 clusters and here's my network config:

- Management + Storage: 2 NIC LACP
- Guest: 2 NIC LACP
- Public: 2 NIC LACP

All the "2 NIC" goes to a 2 stack switches with LACP configured.
Now, here (
http://support.citrix.com/servlet/KbServlet/download/38321-102-714737/XenServer-6.5.0_Administrators%20Guide.pdf)
I see:

[...]Management interfaces

You can bond a management interface to another NIC so that the second NIC 
provides failover for management traffic. Although configuring a LACP link 
aggregation bond provides load balancing for management traffic, active-active 
NIC bonding does not.[...]

Ok, great, this is exactly what I need to have.

My issue is: in all XS I see only one NIC to make traffic during Storage 
Migration, I'm expecting that with LACP I see both NIC to make traffic, but 
this does not happen.

Do you have any experience on this topic?
Since we're migrating to XS6.5 adding a new cluster and moving VM, do you have 
a better NIC configuration to suggest?

Thank you very much!

Reply via email to