problems with EN2511CD / NL2511CD

Jouni Malinen jkmaline at cc.hut.fi
Tue Dec 3 01:26:56 EST 2002


On Mon, Dec 02, 2002 at 10:00:13PM -0800, dean gaudet wrote:

> Dec  2 20:57:23 mewlark kernel: wlan0: 00:40:96:5a:5d:ce auth (alg=0 trans#=1 stat=0 len=6fc=00b0) ==> 0 (authenticated)
> Dec  2 20:57:23 mewlark kernel: wlan0: 00:40:96:5a:5d:ce reassoc (len=32 prev_ap=00:03:2f:02:97:46) => 0(0) (OK, new AID)
> 
> those go on for a while, but the cisco never manages to associate.
> (i tried a WPC11 as well, no luck either.)

As far as Host AP driver is concerned, that station has been
successfully associated. However, you could test this with the latest
CVS version to check whether the station is really acknowledging that
last reassociation response. 2002-10-12 release did not yet do this kind
of verification for the ACKs. The latest CVS version requires ACK before
marking station as associated. It should show both the incoming frame
and the ACK, so you should see two messages about authentication and two
about association (if everything goes well).

If the client card is not reporting any errors, I would just have to
guess that it did not like the reassociation reply for some reason (most
probably it arrived after too long delay). Since the client card sent
reassociation request, I would assume that it accepted authentication
frame from the AP (although, I've seen some stations not caring about
authentication that much..).

> Socket 0:
>   product info: "INTERSIL", "HFA384x/IEEE", "Version 01.02", ""
>   manfid: 0x0156, 0x0002
>   function: 6 (network)

> those look like identifiers for a prism2 card but this one should
> be a prism2.5.

I have seen so many re-uses for PC Card manfids with Prism2/2.5 cards
that nothing seems to suprise me anymore.. ;-)  I wouldn't worry about
this, at least before checking what happens to response frames from AP.

-- 
Jouni Malinen                                            PGP id EFC895FA



More information about the HostAP mailing list