05-03-2012 02:59 AM - edited 07-03-2021 10:05 PM
In the H-Reap Design and Deployment Guide, there is a limitation specified around using H-Reap with voice as follows:
"Roundtrip latency must not exceed 300 milliseconds (ms) for data and 100 ms for voice and data between the access point and the controller, and CAPWAP control packets must be prioritized over all other traffic."
I can't find any explanation of why this limitation is defined.
Is it related to CAPWAP control traffic?
Does it apply only to centrally switched voice traffic, or also to locally switched traffic?
Does anyone have any more detail on why this limit exists?
Does it only apply to Cisco voice handsets (or other clients like Skype, Lync etc.)
Regards
Nigel.
05-03-2012 04:12 AM
The round trip latency for voice is supposed to be no greater than 150ms. So with the capwap tunnel in place you want 100ms so that you still have that 50ms on the LAN side. Make sense?
Steve
Sent from Cisco Technical Support iPhone App
05-03-2012 02:15 PM
So, it is a best design practise in general and not related to specific H-REAP limitations? If so, why is the 'must' statement in there as surely this is all dependant on a 'must' on the LAN connection to callmanager or whichever voice endpoint and back.
Also, if the values vary but still maintain a roundtrip value better than 150ms, this is all slightly pointless, no?
Rocky
05-03-2012 02:18 PM
Correct, it's a best practice. If you exceed the 300ms, you might have some issues with the AP staying 'connected' to the WLC.
Steve
05-03-2012 11:46 PM
Stephen, thanks for the response, but it still doesn't quite make sense top me.
Why should the actual traffic type (which is being switched locally at the AP end of things) determine the difference in latency recommendation?
Whether the traffic is voice or data makes no real difference to the control traffic that passes between the AP and WLC?
The only think I can think of is perhaps something around roaming for voice clients which may generate control traffic if you didn't have H-REAP groups defined?
I understand the over-riding latency limit for the CAPWAP traffic between an AP & WLC (300mS), but why should that latency requirement be any more stringent for voice traffic, when the voice traffic is switched locally and won't hit the link between the AP & WLC?
Regards
Nigel.
05-04-2012 03:50 AM
With flexconnect you don't have to locally switch all of the SSID, in te case where you aren't locally switching the voice is where this would come into play. The doc could be a bit better on breaking this out.
Steve
Sent from Cisco Technical Support iPhone App
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide