Don’t Sweat the Small Stuff: It’s Okay to Mix Cisco Access Points
During the past few weeks, there’s been a lot of chatter regarding FCC 14-30. In early June 2016, the FCC published FCC 14-30 Order, which now allows the use of three additional channels (120, 124, and 128) as well as other power adjustment and updated DFS regulations. Cisco’s compliance with new rules requires the assignment of hardware to a regulatory domain, indicating which rules the device complies.
The new –B regulatory domain was designed to take the place of –A so that access points will be compliant with the FCC 14-30 Order. Per the FCC order, access points shipped before June 1, 2016 are grandfathered to adhere to –A requirements. Access points shipped after the June 1 date must meet the –B requirements. This includes changes in DFS detection requirements, which must re-certify according to the new rules.
Cisco’s full response can be viewed here.
I’ve been hearing from a lot of customers who don’t want to mix the two regulatory domains in their networks. But there simply is no reason not to do so.
You won’t get an argument from me, there are a few nice things in a –B access point that aren’t available in an –A access point. Things that we all want: more 5 GHz channels and more power in U-NII 1. However, other than those minor changes, the –B and –A access points are the same. There are absolutely no operational issues in running both –A and –B on the same controller or controllers.
For example, let’s say you’re adding some access points or building out a new area – and you mix –A and –B access points, what will happen in terms of operations?
If the new channels aren’t added to the DCA list—and by default, they are not—they will not be assigned to any of the –B capable access points. If the new channels are added to the DCA list, it will only be assigned to the –B access points. This won’t be a problem as the clients will still use 120, 124, 128 where it’s available.
I have been in this industry for a long time and I’m not aware of any clients that support U-NII 1, 2, 3 that have failed in these channels. If, for some reason this does happen, simply remove the channels. No harm, no foul.
That leaves different allowed transmit (TX) power. There will be no issues here either – since TPC still works the same as it always has. The Cisco access point product line still operates with a mix of allowed powers in the 5 GHz UNII bands under the –A rules. The Neighbor Discovery Protocol is normalized for this reason and Radio Resource Management (RRM) works just fine. There are no known issues with mixing –B and –A radios in the same air on the same controller. If a user wants to stay consistent about power implications, simply set TPC Max to enforce max power to –A globally and in RF Profiles when in use.
In the last sentence I talked about consistency, and there are a lot of customers—myself included—that just don’t like the idea of a mixing anything. I go so far as to stay away from milkshakes and just eat ice cream cones, when I want a cool treat. But the reality is this is not like mixing Cisco Aironet 1130 and Aironet 3700 Access Points in the same room. Once your –A and –B access points are plugged in and running no one would ever know that a mix existed.
We don’t have operational hiccups absorbing this change like some of our other competitors. At Cisco, we’ve been running mixed environments in Alpha production networks since the Cisco AP 1810 was in development—and that’s a fair amount of time to find irregularities and observe errors. Cisco Mobility Express is built on a –B access point, and supports the –A access points. These devices have been tested over thousands of hours and work as expected.
There is really no reason other than just the perception of a mix to be worried. With that being said, I’m heading down to the ice cream shop and enjoying a frosty milkshake. I hear that they’re really great!