Re: Corporate Networks & Hidden SSID

[Dan Williams <dcbw redhat com>]

On Fri, 2007-01-26 at 12:37 +0100, Grant Williamson wrote:
> Dan,
>     I checked out the cvs code, don't see this change in there yet, are 
> u sure it will make it in 0.7? Also any idea if there are any plans to 
> backport Leap to 0.6.x i.e. for rhel5?

Leap is in 0.6.5 in SVN already, just waiting on the 0.6.5 release.  It
will only work with drivers that support LEAP through wpa_supplicant,
which rules out the airo driver.  0.6.5 is API compatible with what's
currently in RHEL5, so there is a chance that RHEL5's NM could be revved
to include LEAP.  If you're interested in NM+LEAP for RHEL5, I'd suggest
filing a bug in RH bugzilla so we don't lose it.

Cheers,
Dan

> Dan Williams wrote:
> > On Thu, 2007-01-25 at 20:10 +0100, Grant Williamson wrote:
> >   
> >> Dan,
> >>     testing on our corporate eap/tls systems, i.e. what I originally 
> >> tested on, setting this to AP_SCAN 1 it connects.
> >> Why is the default in most cases then AP_SCAN 2, just trying to figure 
> >> out problems it will cause switching it to 1.
> >>     
> >
> > It should only be set to 2 when:
> >
> > 	 * - The wireless network is non-broadcast or Ad-Hoc
> > 	 * - The wireless driver does not support WPA (stupid drivers...)
> >
> > in your case, since the net is hidden, NM will do what wpa_supplicant
> > suggests, and set ap_scan=2.  There's a note in the wpa_supplicant
> > config file that says that if you're using ap_scan=2 then you _MUST_
> > match the pairwise and group cipher parameters to exactly what the AP is
> > set to.  That's the bit NM doesn't allow you to set right now, which
> > will change in 0.7.  So if your AP isn't set to "TKIP CCMP" for both of
> > those then it may not work with NM.
> >
> > Dan
> >
> >   
> >> Matthew Saltzman wrote:
> >>     
> >>> If it's not too much trouble, would y'all mind stripping that long log 
> >>> excerpt from the bottom of future messages in this thread?
> >>>
> >>> Every message is producing a digest now, which kind of defeats the 
> >>> point of getting the digest...
> >>>
> >>> Thanks!
> >>>
> >>>       
> >
> >
> >   
>