Controllerless Networks

last person joined: yesterday 

Instant Mode - the controllerless Wi-Fi solution that's easy to set up, is loaded with security and smarts, and won't break your budget
Expand all | Collapse all

Dropping clients

This thread has been viewed 13 times
  • 1.  Dropping clients

    Posted Aug 28, 2014 02:30 PM
      |   view attached

    I have a weird situation.

     

    The setup.

     

    3 IAP-135's

    Factory_restored

     

    The issue. Clients get dropped, apperently when the AP decides to move them to another AP.

     

    The weird part is devices like a iPad will not reassociate unless you manually go choose it again. Its like the access point told the iPad to not even try connecting.

     

    I attached the config

    Attachment(s)

    zip
    instant5.zip   1 KB 1 version


  • 2.  RE: Dropping clients

    Posted Aug 28, 2014 02:31 PM

    Current firmware



  • 3.  RE: Dropping clients

    Posted Aug 28, 2014 02:36 PM
    Do you have client match enabled?


  • 4.  RE: Dropping clients

    Posted Aug 28, 2014 02:54 PM

    I had the issue with client match on and client match off.

     

    I factory restored, then just created 1 network. Just SSID/pass

     

    Very simple config.

     

    2 of the IAP-135's came from another system and I used the factory reset button to reset them before adding them to the system that already had a perfectly functioning IAP-135 as a controller. The issue began after i added the 2 additional access points. They all show up and everything works fine EXCEPT clients get dropped and do not automatically reconnect.



  • 5.  RE: Dropping clients

    Posted Aug 28, 2014 04:08 PM

    Do you mind sharing the output of "show tech-support"?  You could also open a ticket with Aruba TAC.



  • 6.  RE: Dropping clients

    Posted Aug 28, 2014 05:28 PM
      |   view attached

    dump attached.

    Attachment(s)

    txt
    dump1.txt   801 KB 1 version


  • 7.  RE: Dropping clients

    Posted Feb 12, 2015 06:27 AM

    Was this ever resolved? We're seeing the exact same issue with our implementation.