Cisco DNA Center 0 day plug and play snmp

[pdmarshall]

I am trying to deploy Catalyst 9300 switch with Plug and Play, Day 0 template.

I have been following this blog from Adam Radford

https://blogs.cisco.com/developer/dna-center-pnp-5

While claiming the device I am getting following error:

"SNMP Credentials from Network Settings should not be empty for the selected site."

I can do device discovery so I know the snmp is working, what am I missing please.

I've attached screen shots.

Thanks

[aradford]

Hi,  that screen shot just shows write credential, do you also have read?

[pdmarshall]

Good Morning Adam

Thanks for your reply. Yes, read was configured, but when I clicked on the save icon, it has let me move onto the planned stage.

Unfortunately it is now showing as not contacted. 'file attached.'

I have tried setting the interface config as a switchport access vlan, switchport mode trunk but it is still not connecting.

I have tried putting the cable into the front interface ports as well as the management port.

Any chance you can help again.

Many thanks

Phil

[aradford]

Did you remove the config from the pnp device and reboot it?
That should start the pnp process..

In general, the steps to trouble shoot are:

1) remove config on the pnp device and reboot ("write erase"  and then "reload")
2) wait for the device to boot, DO NOT press enter on the console else you will stop the PnP process
3) after the device boots, check the interface on the upstream device to make sure you can see the PnP device and it has IP address ("show cdp ne de"), check DHCP server that it has allocated an address.  If this step fails need to verify why there is no ip connectivity. Generally connectivity issues are to do with vlan, trunking, port channel etc.. This post might be helpful https://blogs.cisco.com/developer/dna-center-pnp-day-0
4) verify the pnp discovery process completed on the pnp device.  "show run | inc pnp".  Make sure there is a pnp profile

If you have issues at step 3, need to resolve and go back to step 1

[pdmarshall]

Morning Adam

Thanks for the troubleshooting steps, I am not in the office till Monday, but I will have another go then.

Though this is fresh out of the box, no config 9300.  But checked the uplink port and it is in an error-disabled state. I've done a shut no shut a few times and nothing has changed, but I wont be able to restart the 9300 until Monday.

Thanks again

Phil

[aradford]

Ok. So, looks like something with basic networking.

Can you share versions of code for the 9k.

Also can you share the config of the upstream interface? What type of device is it?

[pdmarshall]

Morning Adam

The new device is a C9300-48U, which is connecting to a C9300-48UXM currently on 16.9.5, but after this weekend it will be on 16.12.03a.

I've tried in on both of the following ports with no luck, but again, I've not had the chance to reboot it, which I will do on Monday.

The uplink config are below. and it is also has

license boot level network-advantage addon dna-advantage

Te2/0/40 #DNAC Build# err-disabled 325 auto auto 100/1000/2.5G/5G/10GBaseTX

interface TenGigabitEthernet2/0/40
description #DNAC Build#
switchport access vlan 325
switchport mode access
switchport voice vlan 425
switchport port-security maximum 2
switchport port-security violation restrict
switchport port-security aging time 2
switchport port-security aging type inactivity
switchport port-security
device-tracking attach-policy IPDT_MAX_10
trust device cisco-phone
macro description cisco-phone
auto qos voip cisco-phone
et-analytics enable
spanning-tree portfast
spanning-tree bpduguard enable
service-policy input AutoQos-4.0-CiscoPhone-Input-Policy
service-policy output AutoQos-4.0-Output-Policy
ip nbar protocol-discovery
end

interface TenGigabitEthernet1/0/46
description #DNAC Build2#
switchport mode trunk
et-analytics enable
end

Many thanks

Phil

[aradford]

bpdu guard will shut down the port.  the nw switch will always send spanning tree bdpu as a protection mechanism.  That explains err-disable.

second port looks ok.  The only thing to check is how DHCP is going to get through?   Are you using vlan 1?

Do you have DHCP for it?

[pdmarshall]

Afternoon

DHCP is set up, though Vlan1 isn't and I have been told that vlan1 is a security risk, as such vlan999 was set up for a project for Prime and I should use that.

But I am unsure how the switch is going to respond as i believe it only searches for vlan out of the box.

I have set the config for vlan999 but it still shows in DNA as not connected which isn't a surprise.

Is there a workaround not using Vlan1.

Many thanks

Phil

interface Vlan999
description PI ZERO TOUCH Configuration
ip address 172.20.200.254 255.255.255.0

interface TenGigabitEthernet2/0/40
description #DNAC Build#
switchport access vlan 999
ip nbar protocol-discovery
end

[aradford · Accepted solution]

You need “pnp startup-vlan 999” on the upstream switch.

https://blogs.cisco.com/developer/dna-center-pnp-day-0