V2110 Controller Software Version 9 / Extreme 3825 AP Connection Problems with Windows Clients

  • 0
  • 2
  • Problem
  • Updated 2 years ago
  • Solved
Hi,

we have a problem with the following environment:

WLAN-Controller: V2110 Software Version 09.15 
AP: 3825i
SSID Mode: Bridged@AP

Clients: Windows 7 (with Intel AC 7260 wireless card)
             Apple MAC Book Air

So the problem is that windows clients can connect to the SSID, get an IP Address but most of traffic (tested with ping) gets loss.
On the other side the Apple Device work properly, absolutely no ping loss with that device type.

I' have logged the AP there is a warning message (every second for the windows device)

 bridge.c:660-bridgeUpdate()-AC:FD:CE:E7:0E:96 no longer with us! in port=eth0 ...

What does that mean?

Greetings

Ronny
Photo of Ronny Engelhardt

Ronny Engelhardt

  • 300 Points 250 badge 2x thumb

Posted 3 years ago

  • 0
  • 2
Photo of Ronald Dvorak

Ronald Dvorak, Embassador

  • 45,132 Points 20k badge 2x thumb
I can't answer what the error mean but...

- is the latest WLAN driver installed on the Win client ?
- what device is the target for the ping ?

-Ron
Photo of Hartmut Sachse

Hartmut Sachse

  • 2,598 Points 2k badge 2x thumb
Hello Ronald,

i will add you some more detailed informations, because is the environment of one of our companies customers.

Infrastructure: VSphere/ESXi 6 (not official supported by V2110)
V2110: V9.15.0010.x (first test was with lastest 9.21.0003.x)

Most problematic clients:
 
- Windows 10 notebook with Intel AC 7260 wireless card, latest driver 18.11.0000.2944  (installed last week)
- Windows 10 notebook with Intel AC 7260 wireless card, driver 18.12.0001.3121
- Apple Mac Book Air - Mac OS X El Capitan

Other working clients: iPhone 4s last 7.x iOS, iPhone 6 iOS 9.0.2, some older Notebooks with Intel Wifi  802.11bgn Standard, one with 802.11an

Destination of the ICMP Ping was in most cases the gateway of the subnet and some test with routing instance included (dest. in other vlan/subnet)

---------------
One of the problematic Win7 notebooks (first in listing) showed a unique symptomatic:

I activated the station event log and found this massive intra access point roaming events between radio1 and radio 2 on the test access point. Nearly every one or two seconds (MAC address changed for privacy):

10/01/15 15:42:38    EventType[Roam]    MAC[5C:8D:4E:42:7E:FE]        AP[15065462085G0000]    FromAP[15065462085G0000]        BSSID[D8:84:66:13:9F:79]        Details: Inside AC from AP/Radio[1] to AP/Radio[2] VNS[PHONE-VNS]
10/01/15 15:42:11    EventType[Roam]    MAC[5C:8D:4E:42:7E:FE]        AP[15065462085G0000]    FromAP[15065462085G0000]        BSSID[D8:84:66:13:9F:71]        Details: Inside AC from AP/Radio[2] to AP/Radio[1] VNS[PHONE-VNS]
10/01/15 15:42:11    EventType[Roam]    MAC[5C:8D:4E:42:7E:FE]    AP[15065462085G0000]    FromAP[15065462085G0000]        BSSID[D8:84:66:13:9F:79]        Details: Inside AC from AP/Radio[1] to AP/Radio[2] VNS[PHONE-VNS]
10/01/15 15:42:10    EventType[Roam]    MAC[5C:8D:4E:42:7E:FE]        AP[15065462085G0000]    FromAP[15065462085G0000]        BSSID[D8:84:66:13:9F:71]        Details: Inside AC from AP/Radio[2] to AP/Radio[1] VNS[PHONE-VNS]
10/01/15 15:42:08    EventType[Roam]    MAC[5C:8D:4E:42:7E:FE]        AP[15065462085G0000]    FromAP[15065462085G0000]        BSSID[D8:84:66:13:9F:79]        Details: Inside AC from AP/Radio[1] to AP/Radio[2] VNS[PHONE-VNS]

The SSID was a simple WPA2 AES only with PSK. Same effects on open SSID created for testing.

Later i saw in the connected client report, that this client has a really poor signal from ap perspectiv (-90dBm and lower), but the access point was only 1 meter away. (same effect on both bands/radio). So i think this wifi adapter has a special problem.

Other problematic clients have normal a signal strength, but the connection (ping test) are not stable. I would think the problem is cause somewhere in the infrastructure. I did'nt had the time for a deep review of the LAN infrastructure configuration.

Best Regards
Hartmut
(Edited)
Photo of Hartmut Sachse

Hartmut Sachse

  • 2,598 Points 2k badge 2x thumb
The customer told me, that the issues are fixing now. The issues where caused by a misconfiguration in the switching infrastructure.
Photo of Hartmut Sachse

Hartmut Sachse

  • 2,598 Points 2k badge 2x thumb
One thing to remember is that a large amount of wifi problems are caused by (lan) infrastructure problems. Sometimes the customers infrastructure looks like a black box to you - so ask for environment documentation and clear definition of interfaces between Wifi and LAN. Everytime check for other problems before you searching for the cause of issues in the wireless world, this could make your life easier.
Photo of Hartmut Sachse

Hartmut Sachse

  • 2,598 Points 2k badge 2x thumb
Even if the main problem is fixed. Did someone knowsimilar roaming issues in the event logs like i've posted?

I know clients with bad driver implementions continously roaming between two nearby access points, but roaming on the same access point from radio 1 to radio 2 and vice versa is a strange behaviour.
Photo of Chad Ciszewski

Chad Ciszewski

  • 62 Points
Hartmut - being as specific as you can due to supposed NDA's - what was the issue on the LAN - I am seeing this now in some of our environments as well as on a test AP that I am less than 3 feet from.  Thanks!