"wpa_supplicant -B -w" ups the interface but the Debian docs say it shouldn't

Jaime re2823 at safe-mail.net
Fri Mar 17 21:41:38 EST 2006


Hi all.

I am writing to try to find out whether the following behaviour is a bug
or not:
When I run "wpa_supplicant -B -w" against my ndiswrapper interface, the
wpa_supplicant process appears to bring up the interface:

Result of running "ifconfig -s":
Iface   MTU Met RX-OK RX-ERR RX-DRP RX-OVR TX-OK TX-ERR TX-DRP TX-OVR
eth0   1500 0     563      0      0      0   629      0      0 0     
lo    16436 0       0      0      0      0     0      0      0 0     

Then run "/usr/sbin/wpa_supplicant -B -w -i wlan0 -D wext -c 
/etc/wpa_supplicant.conf"

Then run "ifconfig -s" again:
Iface   MTU Met RX-OK RX-ERR RX-DRP RX-OVR TX-OK TX-ERR TX-DRP TX-OVR
eth0   1500 0     569      0      0      0   629      0      0      0
lo    16436 0       0      0      0      0     0      0      0      0
wlan0  1500 0       2      0      0      0     0      0      2      0

The wlan0 remains up until I kill the wpa_supplicant process, at which
point the interface then goes down again. I get this behaviour using
wpa_supplicant 0.4.5 (under Ubuntu 5.10) and wpa_supplicant 0.4.7 (under
Debian Etch).

The reason why I am writing this email is that Debian's wpa_supplicant
package contains a file which states that when wpa_supplicant is run
with the -Bw flags, "the wpa_supplicant daemon will start, and sit
waiting for the name interface to come up" (rather than bringing up that
interface).

So as far as I can tell, either:
a) running "wpa_supplicant -B -w" SHOULD up the interface (in which case
there is a bug in Debian's wpa_supplicant documentation, or
b) running "wpa_supplicant -B -w" SHOULDN'T up the interface (in which
case I've observed a bug in wpa_supplicant's behaviour).

Any comments on what SHOULD happen would be gratefully appreciated...

Thank you, Dave

PS I logged my original Debian bug regarding this behaviour here:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=356144
(Please ignore the ifupdown comments...)





More information about the HostAP mailing list