[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: wpasupplicant udeb



On Wednesday 28 May 2008 20:59:15 Frans Pop wrote:
> On Monday 26 May 2008, Kel Modderman wrote:
> > Trying to associate with 00:13:10:41:7e:1f (SSID='kelnet' freq=2412 MHz)
> > Associated with 00:13:10:41:7e:1f
> > WPA: Key negotiation completed with 00:13:10:41:7e:1f [PTK=TKIP GTK=TKIP]
> > CTRL-EVENT-CONNECTED - Connection to 00:13:10:41:7e:1f completed (auth)
> > [id=6 id_str=kelnet]
> 
> OK. These definitely look useful to have.

Yep. Especially if the pkg-wpa maintainers are responsible for answering to
failure reports from wpa_supplicant in debian installer environment, this is
the kind of information that will be needed for any clue.

> 
> > The debugging may not be useful per default, but it could just be worth
> > keeping so that there is an attack vector for people reporting problems
> > about using a wpa enabled netcfg, so that they be able to manually invoke
> > wpa_supplicant on another terminal to capture output for analysis (eg. to
> > find out why association to desired access point failed). It just depends
> > if this need is valid and able to offset the desire for very small binary
> > size.
> 
> It would be really great if we could keep the standard messages, but drop 
> the lower level ones. But I guess that would require an upstream change.
> Any chance of that do you think?

I doubt that. At least not in this upstream development cycle. Not sure how
sane it would be to do that either, but it cannot hurt for me to ask if it
would be a consideration in the future.

> 
> IMO the size of the udeb is quite reasonable already. Reducing it further 
> would be nice, but is not a hard requirement for implementing WPA support 
> for default installations.
> 
> Note: please do not upload yet.
> One thing I'm not yet certain of is what priority the udeb should have and 
> how it should be loaded. This will need some discussion.

No problems. Thanks for the comments.

Thanks, Kel.


Reply to: