Hello everybody,
I tried to search if this had already been discussed somewhere but I can't find anything.
In our environment we use a Cisco ISE used to authenticate our WiFi clients. As you may know, before the authentification takes place, the device sending the RADIUS request has to be added as a "network device" on the ISE with a shared secret to make sure it is allowed to perform the RADIUS request.
In a traditional Cisco WiFi deplyoment with controllers all we had to do was add all our controllers to this list, which was manageable (<10). Now, is there any best practice on how to do this when we have potentially thousands of Meraki APs sending RADIUS requests to our ISE (since there is no controller anymore in the Meraki world) ?
Any help would be appreciated !
Thanks.
Solved! Go to solution.
This document is quite detailed, I had never seen it before.
That being said it goes a bit quickly on how to configure thousands and thousands of access points :
"Repeat steps 1 through 7 for additional Meraki access points that will be used in the ISE deployment"
I really want to avoid having to manually configure a network device on the ISE everytime we deploy a new AP.
Just add a subnet in ISE, such as 10.x.0.0/16 - rather than individual AP IP addresses. You probably only need to define a single subnet to cover every access point.
One other way you could perhaps tackle this is to combine ISE's bulk import with Meraki's API.
The API call you could use is this one:
GET /organizations/[organizationId]/deviceStatuses
None
curl -L -H 'X-Cisco-Meraki-API-Key: <key>' -X GET -H 'Content-Type: application/json' 'https://api.meraki.com/api/v0/organizations/[organizationId]/deviceStatuses'
Successful HTTP Status: 200 [ { "name":"My AP", "serial":"Q2XX-XXXX-XXXX", "mac":"00:11:22:33:44:55:66", "status":"online", "lanIp:"1.2.3.4", "publicIp":"4.3.2.1", "networkId":"N_1234" } ]