04-10-2012 02:06 PM - edited 03-01-2019 10:21 AM
Hello all,
I'm setting up our UCS for port channeling and i believe we're having switch configuration issues. My supervisor is our interim network guy and isn't too familiar with the LACP protocol. After configuring the switch for LACP and enabling a Port Channel with an ID of 20 on the UCS, I have 1 of 2 interfaces with a Membership of UP and the other interface has a Membership of Individual.
I feel it's a switch configuration issue and I was hoping someone could help or point us in the right direction?
Here is the configuration he has set:
!
interface Port-channel20
switchport
switchport trunk encapsulation dot1q
switchport mode trunk
!
interface GigabitEthernet4/7
switchport access vlan 30
switchport trunk encapsulation dot1q
switchport mode trunk
channel-group 20 mode active
!
interface GigabitEthernet4/8
switchport access vlan 30
switchport trunk encapsulation dot1q
switchport mode trunk
spanning-tree portfast
channel-group 20 mode active
Thanks!
04-10-2012 02:09 PM
You should also enable Port Fast on the upstream port channel. It's on one member, but apply it to the Port-Channel20 interface also.
Once you do this, down and up the Port Channel and member interfaces from UCS. It should come up.
Let us know.
Regards,
Robert
04-10-2012 02:34 PM
We have these two warning messages. The first one I think we can ignore, the second I believe means these ports should no longer be setup for trunk mode?
%Warning: portfast should only be enabled on ports connected to a single
host. Connecting hubs, concentrators, switches, bridges, etc... to this
interface when portfast is enabled, can cause temporary bridging loops.
Use with CAUTION
%Portfast has been configured on Port-channel20 but will only
have effect when the interface is in a non-trunking mode.
Here's his configuration:
!
interface Port-channel20
switchport
switchport trunk encapsulation dot1q
switchport mode trunk
spanning-tree portfast
04-20-2012 06:30 AM
This should be fine.
Since the UCS is a device that operated in End Host Mode by default it acts as if the vNic is a physical cable plugged into the switch.
Therefore the portfast would be enabled for all vlans because the vnics would have different vlans.
Portfast is typically used for switchports connected to hosts and the UCS is doing just that just passing them over less links if that makes sense.
Is the PC still not working? If not when you "connect nxos" on the fabric interconnect and do a "show run interface x/y" for the interface in the port channel what is the fabric interconnect config?
The upstream switch config looks good providing you are not trying to do a vPC. If you are doing a vPC you will need to configure the vPC on the port channel on the upstream switch.
Joey
04-24-2012 04:03 AM
The one item that stands out to me with your configuration are that the interfaces 4/7 and 4/8 are configured for both access and trunk. Removing the switchport access VLAN 30 and adding switchport trunk allowed vlans 30 should help. Do a shut then no shut and you should be okay
Sent from Cisco Technical Support iPhone App
05-27-2012 08:51 AM
You should add trunk to your portfast command and as the previous person said you should remove the access commands from the physical interfaces
Sent from Cisco Technical Support iPhone App
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide