Communication between APs and controlleurs under 100ms

Unanswered Question
Mar 5th, 2007

I heard from a Cisco representative that communication between the APs and the controlleur(s) should be less than 100ms. Did anyone experienced problems by not respecting that restriction?

This restriction would have an impact on our design. Originaly we were planning to host the controlleurs as well as the WCS on our datacenter and have the APs communicate with the controlleurs/WCS via WAN. We would like to avoid having a controlleur for each site since the sites won't need more than 5-6 APs each...

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 4 (1 ratings)
Loading.
rob.huffman Mon, 03/05/2007 - 09:25

Hi Mehdi,

Have a look at the specs in this H-Reap design guide. It may help explain this restriction and the effect it will have on the "Wireless Rollout";

H-REAP WAN Considerations

Because the H-REAP has been designed specifically to operate across WAN links, it has been optimized for such installations. Though H-REAP is flexible when it comes to these remote network design scenarios, there are still a few guidelines that need to be honored when architecting a network with H-REAP functionality.

H-REAPs may not be placed across WAN links any slower than 128 kbps.

Roundtrip latencies between H-REAP and the controller may not exceed 100 ms.

Between the access point and the controller, a minimum of a 500 byte MTU is supported.

In order to ensure that support for this stated latency limitation is in place, it is strongly recommended that between the access point and controller, priority be configured in the intermediary infrastructure to elevate LWAPP control (UDP port 12223) to the highest priority queue available. Without priority placed on LWAPP control, spikes in other network traffic will very likely cause H-REAP access points to frequently shift from connected to Standalone modes as WAN link congestion prevents access point/controller messages (and keep-alives) from being delivered.

Frequent H-REAP flapping causes serious connectivity issues. Without proper network prioritization in place, it may be prudent to place controllers at remote sites to ensure consistent and stable wireless access.

Note: Whether H-REAP is configured to tunnel client traffic back to the controller or not, the LWAPP data path is used to forward all 802.11 client probes and authentication/association requests, RRM neighbor messages, and EAP and web authentication requests back to the controller. As such, ensure that LWAPP data (UDP port 12222) is not blocked anywhere between the access point and controller.

From this good doc;

http://www.cisco.com/en/US/products/ps6521/products_tech_note09186a0080736123.shtml#t7

Hope this helps!

Rob

Mehdi_ab Mon, 03/05/2007 - 10:50

Thx Rob

I guess it would be "safer" to keep the controllers on the sites and the WCS on the NML since we have no QoS on the WAN link.

Do you know of any real word stats that would clarify the bandwidth needed for each controller to communicate correctly with the WCS.

Maybe a late post but someone else might find it useful.

I did some tests with a wan simluator and the accesspoints joined the controllers fine even with 1000ms delay. Problem was when you tried to connect clients, some did connect and some didn't.

So you better stick to cisco's recommendation if you want a wireless net that is actually working.

Actions

This Discussion

 

 

Trending Topics - Security & Network