rtap mode.

Michail Bletsas mbletsas at laptop.org
Sun Dec 9 22:20:35 EST 2007


David,

The rtap interface wasn't supposed to do anything else but sniff...

M.





David Woodhouse <dwmw2 at infradead.org> 
12/09/2007 09:42 PM

To
Michail Bletsas <mbletsas at laptop.org>
cc
libertas-dev at lists.infradead.org
Subject
Re: rtap mode.







On Sun, 2007-12-09 at 20:17 -0500, Michail Bletsas wrote:
> David Woodhouse <dwmw2 at infradead.org> wrote on 12/09/2007 04:34:28 AM:
> 
> > On Sat, 2007-12-08 at 21:25 -0500, Michail Bletsas wrote: 
> > > > What was wrong with 'iwconfig eth1 mode monitor', anyway?
> > > > 
> > > It does not allow sniffing with the interface running.
> > 
> > Ah, right. Still, what we have at the moment doesn't seem to allow
> > sniffing at all, unless I'm just too dim to drive it... :)
> > 
> http://wiki.laptop.org/go/Wireless#Capturing_wireless_traffic_on_the_xo 
> 
> We use that all the time on the XOs 

Thanks; I have that working now. We weren't setting the MAC address on
the rtap interface, so the kernel was refusing to bring the interface up
because its address was invalid. That check for a valid MAC address
doesn't happen in the older kernel that we're using on the XOs.

Still, it doesn't do anything _but_ sniff, afaict. I've actually removed
the bit of code I quoted, which prevents any TX on the eth%d and msh%d
interfaces when we're in monitor mode (as well as cleaning up the set of
4 or 5 nested functions in the TX path) -- but still it doesn't actually
work, and even association fails. Although association is failing a lot
of the time here in 1cc anyway; maybe that's not related to rtap.
Investigating now...

-- 
dwmw2


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.infradead.org/pipermail/libertas-dev/attachments/20071209/86dfcbb1/attachment-0001.html 


More information about the libertas-dev mailing list