Hi again,

I did a new log dump with some more debug. Also I fixed my
application, it was hammering connman a bit too much. Actually connman
returns already powered WHILE powering on, so my app believed in
connman and started to scan. Now it only listens to property changes,
so it wont start to scan too early, which makes the log much cleaner

The issue remains, the interface get stuck disabled... power off and
on cures the situation.

This new paste shows this:



http://pastebin.com/qAH0mHjR



Until it was powered off rfkill was like this:

rfkill list
0: phy0: wlan
        Soft blocked: no
        Hard blocked: no

ifconfig only showed the lo interface, so the interface was down



After I powered off:
# rfkill list
0: phy0: wlan
        Soft blocked: yes
        Hard blocked: no

ifconfig still considered the interface down


After power on again:
# rfkill list
0: phy0: wlan
        Soft blocked: no
        Hard blocked: no

And ifconfig considered the interface up.

//Richard


On Wed, Aug 6, 2014 at 2:50 PM, Richard Röjfors
<richard.rojf...@gmail.com> wrote:
>
> Hi,
>
> I mentioned this briefly on the mailinglist. If connman dies of some reason 
> (crash or I kill it), my init system will restart connman.
> But connman fails to connect again (well it works if I also make sure 
> wpa_supplicant is restarted).
>
> When connman dies I'm connected to wifi, which I expect to be reconnected to 
> when connman is restarted.
>
> A while after the restart my application gets tired and issues a scan, then 
> connman returns NoCarrier, then the application start to toggle power on 
> wifi...
>
> Before connman gets killed rfkill output looks like this:
>
> # rfkill list
> 0: phy0: wlan
>         Soft blocked: no
>         Hard blocked: no
>
> Afterwards:
> # rfkill list
> 0: phy0: wlan
>         Soft blocked: yes
>         Hard blocked: no
>
> And the log output:
> http://pastebin.com/ecyTPitc
>
> Since wpa_supplicant and connman runs in different processes I'm not 
> completely sure if the order is correct if events happens in the two 
> processes at the same second.
>
> Thanks,
> Richard
_______________________________________________
connman mailing list
connman@connman.net
https://lists.connman.net/mailman/listinfo/connman

Reply via email to