OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
From: Attila Weyland (attila.weylandgmx.net)
Date: Fri Jul 13 2001 - 03:25:04 CDT

  • Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

    Hello Jussi!

    Just to warn you: I don't have any good news.

    One month ago I tried to obtain exactly the same
    information.
    After having tried everything (iwspy, SNMP, hacking
    iwconfig) I wrote an e-mail to Jean Tourrilhes, author
    of the wireless tools. Here's his reply:

    On Tue, Jun 12, 2001 at 03:26:13PM +0200, Attila Weyland wrote:
    >
    > -- CUT --
    >
    > My goal is this:
    > I want to measure the link quality of two acces points
    > from different subnets at the same time.

            Not possible yet (no documentation). Would be done with APLIST
    if ever implemented.

    > I had a look at iwspy.

            Nope. Will work only within the cell.

    >
    > -- CUT --
    >

    As you see the lack of information is the source
    for the trouble we have.

    Lucent (Agere now) doesn't even provide the WaveLAN
    MIB AP. Although I found a little tool which does
    read out the information needed.
    But: Gathering the information via SNMP was not
    very reliable for me. Read-out takes too long and
    displays only the client which are currently
    connected.
    Of even more concern is that according to the 802.11
    standard a station can only be connected to one AP
    at the same time. That means you can't get
    simultaneous statisitics for two or more APs.

    As you correctly mentioned the fact that these
    information is really "flying" around can be seen
    with the Site Monitor under Windows.

    When I made my investigations, I started at the
    LINK layer. After reading the driver and the specs
    I was at the MAC layer.

    A probably time intensive and not guarantied to work
    way might be to generate your own wirless "ethernet"
    frames, to initiate a channel sweep and collect some
    beacons which contain the information.
    Of course these must be done in kernel space.

    As the collection of statistics was only the beginning
    of my diploma thesis I have to continue and can not
    spend much more time on it.

    Sorry for the bad news, but what we need, might not be
    available any time soon.

    Regards,

    Attila