HPE Community read-only access December 15, 2018
This is a maintenance upgrade. You will be able to read articles and posts, but not post or reply.
Hours:
Dec 15, 4:00 am to 10:00 am UTC
Dec 14, 10:00 pm CST to Dec 15, 4:00 am CST
Dec 14, 8:00 pm PST to Dec 15, 2:00 am PST
BladeSystem - General
cancel
Showing results for 
Search instead for 
Did you mean: 

Virtual Connect Routing Question

 
KSF
Occasional Contributor

Virtual Connect Routing Question

Please can someone advise how the routing works within VC if you have multiple ports assigned to a server (see sample config below).

VC EXAMPLE CONFIG
Port 1 - DATA1 network (physically patched to bays 1 & 2)
Port 2 - DATA2 network (physically patched to bays 5 & 6)
Port 3 - DATA1 network (physically patched to bays 1 & 2)
Port 4 - DATA2 network (physically patched to bays 5 & 6)

- All these ports are trunked presenting the same set of multiple VLAN's and so full access is available whether using DATA1 or DATA2.
- Each port is mapped to a different physical NIC.
- This approach is taken to provide redundancy across networks (patched to different switches) and NIC's within each server.

So my question is what route would be taken first, port 1 because it is first in the list? Or is it random?

Is there a benefit if I get half the enclosure to use a different order (DATA2, DATA1, DATA2, DATA1)? The idea being to encourage the traffic to be split more evenly across the links and switches. I assume if all servers have the same profile they would all attempt to hit the first network first (?), so in the 'VC Config' example given above DATA1 would be hit hardest before DATA2. Or is VC more intelligent than this?

I am really trying to see if there is any benefit in having different profiles (with different orders), or the same profile across all? Obviously having the same profile keeps it simple, but it may not be the optimum approach?

Thanks
Kevin
1 REPLY
HEM_2
Honored Contributor

Re: Virtual Connect Routing Question

Kevin:

what really dictates which NIC will be used is the server configuration itself. Are you teaming/bonding these NICs together? If so, the teaming/bonding configuration will dictate which NIC is used.

If the server has NICs in different IP subnets than the servers internal IP routing table will determine the path.

Alternating the config on the server profiles will help to provide a better balance of traffic across the 2 networks. But this balance of traffic could instead be acheived by the NIC configuration on the servers. I.E. Bay 1 server has NIC 1 primary, Bay 2 server has NIC 3 primary... etc.