I am attempting to set up Distributed Switch in a new lab but I have a question about how to properly configure the port LAGs.
If I have a collection of servers, each with two uplinks to a physical switch, how should I correctly configure the LACP LAGs? Do all ports on all servers become part of the same LAG, or should I be creating a different LAG for each new host?
For example:
Lets assume 16 servers. Do I create a single LAG on the DSwitch, mapping all 32 physical uplinks to the same LAG, then replicate that configuration on the physical switch? Or should I be creating 16 different lags, one for each pair of uplinks on each server host?
It seems odd to me that I would configure all servers on the same LAG, containing uplinks to different physical hosts.
Aaron, it sounds like you have it figured out. You can think of the LAG configuration on the distributed switch as a profile that get's applied to each host. The physical upstream switch is configured with a LAG for each host.
Hi Friend,
Plz refer below KB :
For more info on configuring DVS end-to-end, Plz ref below resources:
https://www.youtube.com/watch?v=a9tOtYow0HE
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
If it is useful,plz award point as appropriate
Thanks for the response!
I had already read through these documents prior to posting the question. Videos like this have been helpful: https://www.youtube.com/watch?v=VRZJL5QDBCU but only show physical switch config for a single host. Due to lack of knowledge/experience, I'm having trouble visualizing the LACP configuration of the physical switch.
I'm trying different things in my test lab now to get a better understanding of how this all works. The LAG in vSphere seems to be just a template for any associated hosts, not an actual LAG (that part was confusing to me at first). I now believe that while I only have a single LAG created in the vSphere web client, I must still create multiple LAGs on the physical switch (one for each LAG on each host). My original mistake was assuming that I was to create a single LAG (containing all hosts) on the physical switch.
Aaron, it sounds like you have it figured out. You can think of the LAG configuration on the distributed switch as a profile that get's applied to each host. The physical upstream switch is configured with a LAG for each host.
