Re: Re: Problems with btdun branch



On Wed, 17 Feb 2010 12:23:51 -0800, Dan Williams <dcbw redhat com> wrote:
> On Wed, 2010-02-17 at 14:24 +0100, van Schelve wrote:
>> On Wed, 17 Feb 2010 11:55:44 +0000, Bastien Nocera <hadess hadess net>
>> wrote:
>> > On Tue, 2010-02-16 at 15:03 -0800, Dan Williams wrote:
>> >> On Tue, 2010-02-16 at 11:17 +0100, van Schelve wrote:
>> >> > Hi.
>> >> > 
>> >> > I want to tryout the btdun branch to see how the work that is in
>> there
>> >> > works an can be used in the filed.
>> >> > Therefore I compiled the btdun branch of network-manager and
>> >> > network-manager-applet under Ubuntu lucid.
>> >> > 
>> >> > Now, when pairing the bluetooth mobile using bluetooth-wizard
there
>> is
>> >> > a
>> >> > new checkbox:
>> >> > "Access the Internet using your mobile phone (DUN)"
>> >> > When activating this checkbox bluetooth-wizard freezes. When
>> >> > starting
>> >> > bluetooth-wizard from command-line I see the following messages
when
>> >> > trying
>> >> > to activate the checkbox:
>> >> > 
>> >> > ** Message: dun_start: starting DUN device discovery...
>> >> > 
>> >> > (bluetooth-wizard:2663): GLib-GObject-WARNING **: cannot register
>> >> > existing
>> >> > type `BlingSpinner'
>> >> 
>> >> What do you get for:
>> >> 
>> >> grep -r "BlingSpinner" /usr/lib/gnome-bluetooth/plugins
>> >> 
>> >> or
>> >> 
>> >> grep -r "BlingSpinner" /usr/lib64/gnome-bluetooth/plugins
>> >> 
>> >> ?  Maybe we just need to rename it so it doesn't use a common name
and
>> >> is private to NM.  Since the plugins are shared objects we do have
to
>> be
>> >> careful about the types we register with glib.
>> > 
>> > That would probably happen if you don't export the BlingSpinner
symbol
>> > from your plugin, and use it against an older gnome-bluetooth.
>> > 
>> > gnome-bluetooth in master (and the latest release) uses the spinner I
>> > put in GTK+.
>> > 
>> > Cheers
>> 
>> Hm, after rebuilding the packages again I was able to do the pairing
>> successfully.
>> There is an entry in nm-applet for the bluetooth mobile and under this
>> the
>> available
>> connections are listed. Great!
>> 
>> But there are now two new questions:
>> 
>> 1. As it looks to me it is not possible to use an existing nm system
>> connection. Is this correct?
> 
> No, this should be possible as long as the configuration is there.  If
> not, that's a bug.  At this point only the 'keyfile' plugin allows
> bluetooth configuration.
> 

Then it looks like to be a bug. I have created the connection based on the
documentation I found somewhere at live.gnome.org/NetworkManager but maybe
it is wrong?

[connection]
autoconnect = false
uuid = 9ff1bcc4-15da-416f-9dce-ced289a64cdd
timestamp = 0
type = bluetooth
id = Test-BT

[bluetooth]
bdaddr = 00:16:4e:24:ea:2f
name = bluetooth
type = dun

[ppp]
refuse-mschapv2 = true
no-vj-comp = false
refuse-eap = true
require-mppe-128 = false
crtscts = false
require-mppe = false
mru = 0
nobsdcomp = false
refuse-pap = true
refuse-mschap = true
mtu = 0
refuse-chap = false
lcp-echo-interval = 0
mppe-stateful = false
noauth = true
baud = 0
lcp-echo-failure = 0
nodeflate = false

[gsm]
number = *99#
network-type = -1
band = 0
apn = myapn.tda.t-d1.de
allowed-bands = 1
username = itu002
password = itumobil

[ipv4]
dns = 10.22.33.253;10.22.33.249
dns-search = mydomain.de
dhcp-send-hostname = false
ignore-auto-dns = true
ignore-auto-routes = false
never-default = false
method = auto

[serial]
baud = 115200
parity = 110
stopbits = 1
bits = 8
send-delay = 0

The permissions for the configuration file are correct. I also restarted 
nm-applet and NetworkManager but the connection is not shown.

>> 2. To bring up a bt based connection I need to dial it twice. After
>> selecting the connection the first time
>> nm-applet seems to loop but nothing happens. But when I disconnect and
>> then connect again to the same connection
>> it comes up.
> 
>> I have captured some debug output. I don't know if it's a good idea to
>> post these long logs here. Therefore I put
>> it on pastebin:
>> 
>> http://pastebin.com/m5c9ad0f4 for network-manager debug and
>> http://pastebin.com/m4a0be7b3 for modem-manager debug
> 
> The line:
> 
> NetworkManager: modem_prepare_result: assertion `state ==
> NM_DEVICE_STATE_CONFIG' failed
> 
> is a bug in NM.  Need to fix that; I think this got broken due to the
> PIN fixing changes when I merged master last week.
> 
> Dan


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