Showing results for 
Search instead for 
Did you mean: 

DFS testing, how to?

Contributor II
We are currently using only non-DFS channels (36, 40, 44, 48, 149, 153, 157, 161, and 165) in our Ruckus deployment. With only 9 channels to choose from in 5GHz we hard coded our channel width to 20MHz.

At some point in the future we may jump to 40MHz or even 80MHz wide channels which would necessitate using DFS channels. If we did jump all the way to 80MHz wide channels my understanding is this would still be dynamic with clients negotiating the best width on the fly with 80 being the max but clients could choose 40 or even 20.

I want to test whether in our geographic area, DFS is a problem. Here is my plan:
  • Take 4 Ruckus R710 APs and pin them to channels 52, 56, 60, and 64
  • Let them run for a period of time on these channels
  • Look at log files to see if there were channel changes due to DFS
  • Pin the same 4 APs to channels 100, 104, 108, and 112 and repeat the test
  • Pin the same 4 APs to channels 116, 120, 124, and 128 and repeat the test
  • Finally, pin the same 4 APs to channels 132, 136, 140, and 144 and repeat the test
Three questions
  1. Is this a sound test methodology?
  2. How long should each test run? My understanding is a channel change could be caused by weather radar or airport radar. These types of radars operate a lot so I was thinking two full days for each of the 4 tests would be adequate
  3.  Where exactly would I look to see if there was a channel change due to DFS? We use a Ruckus Virtual Smartzone controller. Do we look on the controller or each AP?

Valued Contributor II
hi David,

your test plan is reasonable.
you will run test with 4 AP's with fixed channels for how many days?
Read this doc which gives how, when & where wrt to DFS, Radar & its behavior..

to identify instances of Radar/DFS event, i suggest setup a syslog server so that you can search the syslog with keywords like DFS or radar... that shud do the trick..

best of luck..

Contributor II
I thought about running a test on any one DFS channel for two days. My understanding is a DFS event would be caused by weather radar or airport radar. The airport in my area is 20 miles away straight line distance and of course they use radar every day. Weather radar is used daily as well, at least that is what my perception is watching my nightly weather forecast.

Would two days be a reasonable test length?

New Contributor II
In testing I triggered Radar events on channel 124 and 128.  These are the only two channels that have triggered anything.  Can I still turn DFS on and then un-select 124 and 128?

Valued Contributor II
The thing that burned me recently is that certain helicopters also trigger DFS events. I was DFS free until the local police department started doing training routes that go over one of my sites. So it's not just airports and weather radar though if you live near either of those things there's almost no chance of channels 100-132 being usable.