Interesting case here maybe, which is buggering me since a couple of days.
We have an environment with hundreds of LAPs over multiple sites throughout the globe. We have main controllers in datacenter for remote LAP (H-REAP mode) and to terminate the WLAN for guests (using the mobility anchors). Larger sites with multiple LAPs have local controllers.
With this setup we figured the best practice is to use DHCP to assign controller addresses in the L3 LWAPP Discovery process.
Recently I found that our DHCP administrators are setting up hardcoded DHCP option 43 in order to get it to work. All good, it works. However, I would prefer to use Option 60 with VCI to assign Option 43 values (as I can seein the forseeable future there will be need for smartphones, prnters or other devices requiring other Option 43 values. So, in my minset VCI mapping is the way to go.
Now, I have been working with one of my system admins to set this up correctly, and followed Cisco's document to the T. Checked, double checked, triple checked it. Restarted the DHCP service, restarted the DHCP server, just to rule out any possible issues.
But, packet tracing the DHCP upon LAP boot I do see the server is absolutely not sending any Vendor specific info, not even Option 43. Only when I hardcode Option 43 (so not the sub-coded 241) with the local controller IP address the LAP receives the right info and registers to that controller.
I even have a TAC case opened to request confirmation if the document I have from Cisco is not containing any errors, and TAC confirms this is the correct document.
Anyone ever run into such issue. I'm convinced it must be something wrong on the DHCP servers (Freaky Mickeysoft), but I'm a noob at server management.
Any one any clues what to check?
Thanks in advance,