rt2x00.serialmonkey.com

Hello all,

First and foremost: thank you very much for the effort you put into getting an open source driver out there for our hardware. Also, congratulations on making it into the mainline kernel

I'm having quite a bit of trouble getting my USB wifi dongle to work properly with a WPA encrypted network, and I hope you will be able to give me some advice. I have a dongle that identifies itself as "14b2:3c02 Atheros Communications Inc", and I'm running Ubuntu 8.04, kernel 2.6.24-16-generic. Ubuntu recognizes the dongle automatically, and loads the rt2500usb driver for it. This driver seems to be working nicely with WEP and even WPA for about half a minute, and then comes to a grinding halt when more than half the packets get lost and the speed drops to a few k per second.

I've solved this problem for someone else by installing the rt2570 driver from the CVS, a few weeks ago, which did the trick nicely. That is: for a WEP encrypted network. But now I've done the same thing for my own dongle I've discovered that upgrading the driver to rt2570 makes me lose the capacity to connect to a WPA encrypted network.

I've tried to compile the beta release instead of the CVS release, but this gives me an error:



My question is: why can't I connect to a WPA encrypted network with the rt2570 driver from the CVS (even though I can with the rt2500usb driver), could I perhaps get it to work with the beta release, and if so; how do I get it to compile? Or do I perhaps need to use a different driver altogether?

Thanks for any help in advance,
Tim

Are you sure the rt2500usb is doing the trick? If the device is identified as "14b2:3c02 Atheros Communications Inc" then I would really assume the Atheros driver would be required instead of Ralink driver... The USB ID is however known to be rt2500usb, but with some manufacturers changing the chipset without changing the USB ID, that doesn't say much.

The compile error is simple, you need the kernel _sources_ and not the headers.

_________________
Regards,
Ivo van Doorn
Project Administrator
http://rt2x00.serialmonkey.com

I'm quite sure it's the rt2500usb driver. I've blacklisted that driver now, and the hardware isn't doing anything anymore without a modprobe rt2570 The name surprised me too... It's a Conceptronic device, other than that I can't tell you much.

Do you think it would be useful to install the kernel sources and give the beta driver a try?

Thanks for the quick reply!

I think Ubuntu has a rt2x00 update in backports that should resolve your problem as well.
Otherwise using the legacy rt2570 should also work.

_________________
Regards,
Ivo van Doorn
Project Administrator
http://rt2x00.serialmonkey.com

Thanks, I'll try that now

Which download would that be?

The rt2570 from our download page.

_________________
Regards,
Ivo van Doorn
Project Administrator
http://rt2x00.serialmonkey.com

Inability to handle AES - CCMP - encryption was a long standing problem with the legacy rt2570 driver, which is what kept WPA encryption from working when that was specified - although I believe TKIP encryption was OK. A fix that handles CCMP encryption - at least for adhoc operation - was put into CVS on 5/4/08. So if you download the latest source tarball and build, WPA with AES should now work for you.

If it does not, please compile and run with debug enabled and post the log here.

Thanks,

(edit) at least for WPAPSK.

_________________
Yr Hmbl Obt Svt & c
Bryan - In favor of Big Oil. Big fan of General Grievous.

Hello,

I have a problem that might be related to that - I did not manage to get WPA working with the daily rt2570 CVS snapshot (taken today) on my 2.6.24 gentoo box (kernel 2.6.24-gentoo-r8).
Everything looks fine, but the connection to my AP ("f9") does not succeed. Taking logs in debug mode, there is a line saying
which may indicate the final problem. I am 100% sure that I am using the right key (connection to same AP with a different PC/Wifi device works fine).
The steps that I'm performing to connect are:

It would be veeeeeery kind if someone could check the attached (commented) logfile that has been taken with debug=31 - after spending quite some time on this issue now I believe that this could be problem in the driver.
The device that I'm trying to use is an ASUS WL-167g (the first version, not the one that requires rt73).

Thanks for your help (and - of course - for all your work on the drivers!!)

With today's tarball, everything works - which is strange, considering the very few changes in CVS... but most important is that it works fine now