cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
465
Views
1
Helpful
2
Replies

DNA Center change of IP address fails

lgd1
Level 1
Level 1

Hi,

We are currently moving our DNA Center 2nd gen appliance DN2-HW-APL-L from one building to our DC. This is forcing us to change the IP addressing for the enterprise port, but when we try to make this change via the sudo maglev-config update command, we receive a message saying that the validation failed for the default gateway and the dns server. We are positive that these parameters are correctly configure because we have other devices under the same subnet working with them.

As said before, this appliance is being moved to a DC, connected to ACI leaf.

Is there any way to check logs or verify manually if we are having any configuration problem? I am considering reimaging the appliance at this point as last resort.

 

Thanks

Lgdi.

2 Replies 2

pieterh
VIP
VIP

I assume you have verified the connected switchport configuration allows access to the gateway (vlan / ACL's)?

it may well be the enterprise port is not connected ?
as a result the gateway cannot be validated on this interface

the maglev setup wizard expects the interface layout as below
     The wizard discovers all of the ports on the appliance and presents them to you one by one,
     in separate screens, in the following order:
     a. (Required) 10-Gbps Enterprise Port                  Network Adapter #1   enp94sf0f0
     b. (Required) 10-Gbps Cluster Port                      Network Adapter #2   enp94sf0f1
     c. (Optional) 1-Gbps/10-Gbps Management Port Network Adapter #3   eno1
     d. (Optional) 1-Gbps/10-Gbps Internet Port         Network Adapter #4   eno2

older setup wizards accepted a more free interface layout,
you may have configured another interface than enp94sf0f0 with the enterprise ip address ?
so confirm that adapter#1/enp94sf0f0 is connected and configured with your enterprise address

Hi Peterh,

Thanks for the suggestion, actually the problem was related to the SFP model.

During installation in the new DC, we changed SFPs from 1Gb to 10GB for the Enterprise ports (We were also intending to form a Port-channel between FEX and DNA Appliance). With this intention, we followed Cisco´s guide for installation: https://www.cisco.com/c/en/us/td/docs/cloud-systems-management/network-automation-and-management/dna-center/2-3-3/install_guide/2ndgen/b_cisco_dna_center_install_guide_2_3_3_2ndGen/m_plan_deployment_2_3_3_2ndgen.html

This guide indicates that SFP-10G-SR-S (Short range, MMF) is a supported media type, and we had a couple of them in our bag, so we installed them on both sides. We observed that LED lights were not even turning on, so after some days of troubleshooting (investigating if we could check some logs in the CIMC or trying to force speed to 10Gb) we finally tried replacing the SFPs with SFP-10G-SR that we had to spare (We would tried anything at this point, even SFPs that we thought would not be compatible) And that did it.

After the replacement, interfaces immediately came up and the default-gateway and DNS server errors disappeared. (By the way, we replaced SFPs on the DNA Center side only.. Nexus are still on SFP-10G-SR-S)

I don´t know if this would be the case for everyone trying to use the first kind of media type, but if it happens, then I hope this info helps. Maybe Cisco´s guide should be corrected, we struggled long time until we came to the solution.

Thank you!

Lgdi.