Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

Hyper V SCVMM and the Nexus 1000V Nexus Uplink Nic getting kicked out from VMM

Hello all!

 

Bit of a long post, if you guys only want to read the problem look for the *

 

I'm a network technician student who started my trainee "internship" awhile ago.

I've been tasked to setup and do a proof of concept for the Nexus 1000V switch in a lab environment,

using Hyper V as the virtualization platform.

 

I did not set it up according to Microsofts best practises (i do not have a standalone third host with the AD, SQL, SCVMM services installed on it),

i only have two HP Proliant hosts that are currently in a HA Cluster and those roles have been installed and are currently running in the cluster.

To set it up i installed the second DC server on my laptop using VMware Player, once i got the cluster up and running i installed the first DC and

moved the FSMO services to it.

 

I made a super simple network topology, and the configuration in the Nexus 1000V is also very simple.

In my education we had the cisco routing and switching course, using different versions of IOS.

When i got this assignment i had never used NX OS before so bare with me.

 

The network topology looks like this roughly

A little bit more detailed view

 

Sadly i do not have the configuration right now for the physical switch, i can get it later if you guys request it.

 

*** The problem is this

When adding the Nexus 1000V logical switch to the physical NIC it looks good (i add the veth adapter to it at the same time).

Adding both hosts (one after the other) to the logical switch using the same portprofiles, uplink profiles, corresponing network adapter (same number and manufacturer), same network segment and VM network.  

Both hosts are clustered using the Hyper V extensible switch.

HBGHV1 Heartbeat 172.21.2.10 255.255.255.0

HBGHV2 Hearbeat 172.21.2.11 255.255.255.0

HBGHV1 Live Migration 172.21.1.10 255.255.255.0

HBGHV2 Live Migration 172.21.1.11 255.255.255.0

When looking in the Administration and Fabric tab, the switch and adapter is fully compliant

When the host cluster updates it self suddenly the physical network card dissapears from VMM (Remote Dekstop still visible on Hyper V Hosts)

The logical switch looses its uplink port and the only way for me to get it back is to remove the logical switch and add it again,

when i try to remediate the non compliance i cant because the logical switch needs an uplink port to do that.

 

I saw this error message in the VMM log, do i get this issue because i havent specified any system network? (neither on the ethernet uplink or vethernet adapters)

I will check the  virtual adapters connected to the Nexus Appliance

This is what it looks like when running the show modules command in VSM

I really appreciate the help!

If there is anything else you guys need just ask

 

Best regards

Peter

175
Views
0
Helpful
0
Replies