DNAC Provisioning slow after changing the Management IP to Loopback0

[mjrduarte]

Hi!

As the title says, once I configure the Loopback0 interface and change the Management IP in DNAC to use the Loopback0 interface, operations on the switch become really slow (Resync, provisioning...).
Simply adding the switch to the fabric as a Border and Control Plane node took 2 hours.
Any idea why this is happening?

Thanks in advance for any help.

[rasmus.elmholt]

Hi  @mjrduarte 

Are you sure this isn't a routing issue? Firewall, asymmetric routing.

There should not be any difference.

If you want to test you can just change the IP address after the discovery.

Provision -> inventory -> actions -> edit device -> Management IP.

[mjrduarte]

Thanks.
I didn't discover the device. I provisioned it through PnP.
So at the end of PnP, I have a vlan interface with the IP from DHCP, and a trunk to the upstream switch.
At this point the management IP in DNAC is the vlan interface IP. I then configure a loopback0 interface with a /32 IP and then create routing so that I can reach that IP (the route points to the vlan interface IP). I dunno if I'm doing something wrong.

[Preston Chilcote]

You can save a lot of hassle by removing the DHCP address during PnP in favor of the static management.  This way you don't have to go back and change the management IP manually in Catalyst Center Inventory.  The only caveat is that the new IP will need to be reachable for PnP to succeed.  https://community.cisco.com/t5/networking-knowledge-base/updating-management-ip-address-of-a-switch-during-the-pnp/ta-p/4820793

[mjrduarte · Accepted solution]

I've solved my issue.
I reconfigured my L2 trunks to L3 all the way to the core.
Thank you for the help.