cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1064
Views
10
Helpful
5
Replies

VPC / Cisco ACE and the Nexus 2K and 5K

grichardson661
Level 1
Level 1

Hi all,

So we have a test environment that looks like the following. We have 2 5K's switch 1 and switch 2. Switch 1 has two 10gb connections downstream to a 2K and switch 2 has two 10Gb connections downstream to the other 2K. We have a few servers that are multi-homed with LACP and VPC via the 2Ks and it works a treat.

We have our Cisco ACE 01, ports 1 and 2 going to one of the 2K's and we have ports 3 and 4 going to the other 2K, ACE02 ports 1 and 2 going to one of the 2K's and we have ports 3 and 4 going to the other 2K. If i enable VPC and none LACP based etherchannel i cannot get the ACE's talking to each other, but looking at the VPC status its all healthy and up.

Has anyone managed to multi-home the ACE between two 2K's with VPC successfully? 

If I disable the links so each ACE only has links upstream in a traditional port-channel and not cross connected, the ACE's can see each other with no issues.

Cheers

5 Replies 5

Reza Sharifi
Hall of Fame
Hall of Fame

Hi,

When you multi-home the ACE to 2 different 2ks, you put all 4 interface in the same vPC right? If not, can you try that and test again?

HTH

Hi Reza, thanks for the reply.

Yes all four connections are in the same port-channel and VPC. Any other ideas? VPC is compatible with an ON etherchannel right?

Cheers,

Hi,

Can you try using LACP?

Thanks,

Doh.. so we had a cable patching issue in the end. Let this be a lesson to all networking chaps - always check the basics first! Now we have patched the cables as per design the VPC has been established and works.

Now we  have VPC is working we are simulating link failures. When we restore a shutdown physical port within the port-channel/VPC that sits between the 2K and ACE (simulating a port failure) the ACE's lose sight of each other for about 10 seconds and causes an short outage until the port is up and up. The logs on the ACE show 'the Peer x.x.x.x is not reachable. Error: Heartbeat stopped. No alternate interface configured' but the VLAN for the FT interface is carried over all four ACE NIC's that are multi-homed to two 2K's... very strange, i would not expect this, it's like the MAC addresses for the FT interface are waiting to be timed out on the 2K until they are switched on another interface within the port-channel and VPC.

Anyone seen this before?

I've resolved this by removing carrier delay 30 from the physical ACE interfaces. As the ACE is connecting to the 2K where spanning-tree is not running so there is no requirement for carrier delay on the ACE.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: