[c-nsp] Mobile plant roaming speeds

[Ron M.]

Absolutely agree, but the client also needs to support .11k, and lots of
the industrial client devices don't do that.

On Sat, Jul 28, 2018 at 12:30 PM, Charles Sprickman <spork at bway.net> wrote:

> This is dated (I don’t think Apple is the only one supporting this in
> clients anymore), but 802.11k exists to give the client information about
> the other APs in a network - I’d suspect any client supporting this would
> be less “sticky”:
>
> https://www.cisco.com/c/en/us/td/docs/wireless/controller/
> technotes/5700/software/release/ios_xe_33/11rkw_DeploymentGuide/b_
> 802point11rkw_deployment_guide_cisco_ios_xe_release33/
> b_802point11rkw_deployment_guide_cisco_ios_xe_release33_chapter_010.html
>
> Charles
>
>
> On Jul 28, 2018, at 11:40 AM, Ron M. <ccie4526 at gmail.com> wrote:
>
> Most of this will be dependent upon the client device itself.
>
> I've got a number of mobile client devices in a glass production facility
> running on a WLC5520/IW3702 infrastructure, and my biggest issue is with
> the stickiness of the client device (Moxa). Even with "forced" roaming from
> the WLC, the client would not willingly roam unless a special feature of
> theirs "turbo-roaming" was enabled... and oh by the way, that turbo-roaming
> function is limited to only three channels. Obviously originally designed
> for a 2.4GHz deployment where you only have three non-overlapping channels
> to work with... but in my facility my design was using .11n modulation with
> 20MHz channels on 5GHz so I could spread out the APs and limit co-channel
> interference. And we discovered the Moxa client wouldn't aggressively roam
> to a new AP even down at -80 to -85dB RSSI... unless we had their
> "turbo-roaming" enabled, which of course defeats the purpose of using the
> additional channel space available at 5GHz.  When I took all this back to
> Moxa support, their reply was essentially "that's how it is, if you want
> the code changed, submit a feature request and pay for it." So... the
> immediate fix was to limit the AP infrastructure to only using 3 channels.
> The long term fix is that as the Moxa radios break, they'll be replaced
> with something non-Moxa. And we won't ever be buying Moxa client devices
> for any of our other facilities. Their attitude = loss of customer.
>
> So, word to the wise, look VERY closely at the capabilities AND limitations
> of the client devices. Test them thoroughly before issuing the purchase
> order.
>
> On Sat, Jul 28, 2018 at 2:32 AM, Mal via cisco-nsp <
> cisco-nsp at puck.nether.net> wrote:
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>
>
> ---------- Forwarded message ----------
> From: Mal <malz at jetlan.com>
> To: <cisco-nsp at puck.nether.net>
> Cc:
> Bcc:
> Date: Sat, 28 Jul 2018 18:02:44 +0930
> Subject: Mobile plant roaming speeds
> Interested if anyone having success using 802.11n/ac in mine wifi
> deployments with mobile clients (trucks/plant), roaming speeds in the
> order of 15-30kmh ?
>
> Broad question, given the roaming responsibility of the client device.
>
> Mal
>
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>
>
>