Issues with Clients moving from Autonomous to LWAPP

[da.beaver]

I have a fairly large wireless rollout of access points at my hospital's main campus and then I have 41 remote sites that have approximately 3 to 10 access points at each location.

My problem is this, I just converted one site from 9 autonomous access point over to LWAPP. The site has a 10 meg metro Ethernet connection so I know its not the link back to me. We have 26 Motion Tablets with Atheros wireless clients at that location that utilize a Citrix application called Logician. We have been having some issues with the Motion tablets holding the Citrix connection, but once we converted this site, those clients have really been having issues.

There is a high density of access points so coverage is not an issue, but I didn't know if anyone else out there may have a similar scenario and could lend some advise as to why the LWAPP causes so many issues.

Thanks in advance.

[jwadleigh]

Hi. LWAPP itself should not cause service to be worse; because of the dynamic nature of its settings, it should actually improve connectivity.

Do these wireless clients have access to any services (e.g. internet) that do not require use of Citrix? If so, how is the performance there? If you're getting consistent, reasonably quick internet access, but your Citrix session keeps failing; it may not be a WLAN issue per se.

I have done several implementations of Citrix over WLAN for hospitals. Most of the time if there are connection interruptions, it boils down to one of two problems.

1.) Interference: Hospitals are very hostile environments for RF. Various building materials, large machinery, and irregular shapes in the floorplan make it difficult for RF to travel consistently.

2.) Citrix settings: Citrix is a highly configurable application. There are session timeout settings which can be tweaked which may make the system more tolerant to momentary lapses in connectivity.

My suggestion would be to see if the issue is limited to Citrix traffic only; and if it is, look at how Citrix is configured. If you're finding sub-par performance with all wireless traffic, then maybe it's an interference issue that keeps knocking you off the WLAN. If that is the case, I would suggest a professional wireless consultant should come in and decide on the best methods for giving you more consistent service.

Hope this helps! If so, please rate.

[wififofum]

I respectfully disagree that Cisco's dynamic features always improve connectivity. Auto-RF Dynamic Channel Assignment disassociates clients before making changes. If the client has other APs in the vicinity, it can survive it, assuming the app is tolerant of delay. In the hospital environment, device usage can be very inconsistent and intermittant as compared to a corporate user. These usage patterns suffer most from the controller's inherent association managemt behavior. The controller will disassociate clients if no traffic is seen over the default timeout intervals. Therefore, consider the WLAN Session Timeout and Controller Client Activity Timeout as well. Persistent Connection may increase the client traffic, which in turn will keep the client associations active.

[da.beaver]

Thanks for your response. My WLAN Session Timeout is set to 28800 seconds. I'm not sure where the Controller Client Activity Timeout is. Can you tell me where to find that?

Thanks.

[wififofum]

Sorry - I was referring to the User Idle Timeout (seconds) under the Controller Tab. You may also want to extend the ARP Timeout as well.

[Scott Fella]

Hard set the channels and find out if the preamble needs to be set to long or short. Try to configure the lwap the same as how you had it in autonomous and see how the clients react.

[da.beaver]

Thanks to both of you for your reply. I have to agree with jwadleigh. I don't really have issues with any other application other than Citrix. We have had the laptops and tablets setup a constant ping back to the core switch and never dropped a packet, while all along the Citrix app was dropping like mad. We also monitored the 100 meg ethernet pipe to this location to see if there were any drops or lost packets and nothing there either. It seemed that most of our issues were a result of the Motion Tablets, but with the rollout of a couple of laptops, the issues are continuing to surface. I have also set the channel on each access point as well and the same results continue.

I think we have a Citrix issue...

[Scott Fella]

Make sure you have persistent connection set on Citrix. There are many causes for Citrix to loose connection in an hospital environment.

[wififofum]

Hey Fella,

Do you have any links describing how to configure the persistent connection feature in Citrix?

Thanks,

[Scott Fella]

Here is a link from one of our Citrix engineers.

http://support.citrix.com/article/CTX104147

This will help, but will not solve any issues due to poor ap placement, configuration, interference and or user error.

Hope this helps.

[dennischolmes]

Hey Dave,

Just cruising the forums and saw your post here. Let me know what you discover about the error from the dbase when you get the chance. I suspect that is our issue.