GNOME.org
 

Re: System settings with keyfile

Dan Williams wrote:

On Mon, 2008-07-28 at 16:55 +0100, Robert Piasek wrote:

Dan Williams wrote:

On Mon, 2008-07-28 at 15:04 +0100, Robert Piasek wrote:

Dan Williams wrote:

 On Mon, 2008-07-28 at 13:15 +0100, Robert Piasek wrote:
Dan Williams wrote:
| On Sun, 2008-07-27 at 23:02 +0100, Robert Piasek wrote:
|> On Sunday 27 July 2008 22:40:25 Dan Williams wrote:
|>> On Sun, 2008-07-27 at 22:32 +0100, Robert Piasek wrote:
|>>> Hi Dan,
|>>>
|>>> The example you've sent me was great. I had to adjust it to WPA, but
|>>> after I got ssid part correctly it wasn't a problem. During my
testing I
|>>> observed some strange timeouts.
|>> When the supplicant doesn't connect, there are usually driver or config 
|>> issues.  Can you add "-dddt" to the end of the Exec= line in
|>> your
|>>
/usr/share/dbus-1/system-services/fi.epitest.hostap.WPASupplicant.service
|>> file, then "killall -TERM wpa_supplicant" and reproduce the association |>> failures? Then grab the output of /var/log/wpa_supplicant.log and mail 
|>> that to me.
|>
|> When everything is as it should be, log looks like:
|>
|>
|> Jul 27 22:53:52 [kernel] [29721.772205] wlan0: deauthenticate(reason=3)
|> Jul 27 22:53:52 [kernel] [29721.787568] mac80211-phy0: failed to
remove key
|> (0, 00:1c:10:e9:50:b8) from hardware (-22)
|> Jul 27 22:53:54 [kernel] [29724.165914] ACPI: PCI interrupt for device
|> 0000:03:00.0 disabled |> Jul 27 22:53:54 [kernel] [29724.370012] PM: Writing back config space on 
|> device 0000:03:00.0 at offset 1 (was 100102, writing
100106) |> Jul 27 22:53:54 [kernel] [29724.572766] Registered led device: iwl-phy0:radio |> Jul 27 22:53:54 [kernel] [29724.572817] Registered led device: iwl-phy0:assoc |> Jul 27 22:53:54 [kernel] [29724.572859] Registered led device: iwl-phy0:RX |> Jul 27 22:53:54 [kernel] [29724.572903] Registered led device: iwl-phy0:TX |> Jul 27 22:53:54 [kernel] [29724.592741] ADDRCONF(NETDEV_UP): wlan0: 
link is
|> not ready |> Jul 27 22:53:56 [kernel] [29726.281451] wlan0: Initial auth_alg=0 |> Jul 27 22:53:56 [kernel] [29726.281451] wlan0: authenticate with AP |> 00:1c:10:e9:50:b8 |> Jul 27 22:53:56 [kernel] [29726.284160] wlan0: RX authentication from |> 00:1c:10:e9:50:b8 (alg=0 transaction=2 status=0) |> Jul 27 22:53:56 [kernel] [29726.284170] wlan0: authenticated |> Jul 27 22:53:56 [kernel] [29726.284176] wlan0: associate with AP |> 00:1c:10:e9:50:b8 |> Jul 27 22:53:56 [kernel] [29726.287316] wlan0: RX AssocResp from |> 00:1c:10:e9:50:b8 (capab=0x411 status=0 aid=2) |> Jul 27 22:53:56 [kernel] [29726.287325] wlan0: associated |> Jul 27 22:53:56 [kernel] [29726.287353] wlan0: switched to short barker |> preamble (BSSID=00:1c:10:e9:50:b8) |> Jul 27 22:53:56 [kernel] [29726.309101] ADDRCONF(NETDEV_CHANGE): 
wlan0: link
|> becomes ready |> 
|> All takes 4 sec and NM is happy.
|>
|>
|> When something is wrong, it looks like:
|>
|> Jul 27 22:54:07 [kernel] [29737.306456] wlan0: deauthenticate(reason=3)
|> Jul 27 22:54:07 [kernel] [29737.327661] mac80211-phy0: failed to
remove key
|> (0, 00:1c:10:e9:50:b8) from hardware (-22)
|> Jul 27 22:54:09 [kernel] [29739.915016] ACPI: PCI interrupt for device
|> 0000:03:00.0 disabled
|> Jul 27 22:54:09 [kernel] [29739.915016] ACPI: PCI Interrupt
0000:03:00.0[A] ->
|> GSI 17 (level, low) -> IRQ 17
|> Jul 27 22:54:09 [kernel] [29739.915016] PM: Writing back config space on 
|> device 0000:03:00.0 at offset 1 (was 100102, writing 100106)
|> Jul 27 22:54:09 [kernel] [29740.240980] Registered led device:
iwl-phy0:radio
|> Jul 27 22:54:09 [kernel] [29740.240980] Registered led device:
iwl-phy0:assoc
|> Jul 27 22:54:09 [kernel] [29740.240980] Registered led device:
iwl-phy0:RX
|> Jul 27 22:54:09 [kernel] [29740.240980] Registered led device:
iwl-phy0:TX
|> Jul 27 22:54:09 [kernel] [29740.268428] ADDRCONF(NETDEV_UP): wlan0:
link is
|> not ready
|> Jul 27 22:54:33 [kernel] [29763.757816] wlan0: Initial auth_alg=0
|> Jul 27 22:54:33 [kernel] [29763.758142] wlan0: authenticate with AP
|> 00:1c:10:e9:50:b8
|> Jul 27 22:54:33 [kernel] [29763.761985] wlan0: RX authentication from
|> 00:1c:10:e9:50:b8 (alg=0 transaction=2 status=0)
|> Jul 27 22:54:33 [kernel] [29763.762450] wlan0: authenticated
|> Jul 27 22:54:33 [kernel] [29763.762810] wlan0: associate with AP
|> 00:1c:10:e9:50:b8
|> Jul 27 22:54:33 [kernel] [29763.766833] wlan0: RX AssocResp from
|> 00:1c:10:e9:50:b8 (capab=0x411 status=0 aid=2)
|> Jul 27 22:54:33 [kernel] [29763.767447] wlan0: associated
|> Jul 27 22:54:33 [kernel] [29763.767481] wlan0: switched to short barker
|> preamble (BSSID=00:1c:10:e9:50:b8)
|> Jul 27 22:54:33 [kernel] [29763.789854] ADDRCONF(NETDEV_CHANGE):
wlan0: link
|> becomes ready
|>
|> 25 sec and NM times out (after 15?).
|
| Yeah, lets do the "-dddt" thing and then
| grab /var/log/wpa_supplicant.log, because that 25 second interval is
| probably the supplicant scanning for the AP but not finding it for some
| reason.
|
| Dan

Let's see:

1217246610.187586: State: DISCONNECTED -> DISCONNECTED
1217246610.187591: wpa_driver_wext_set_operstate: operstate 0->0 (DORMANT)
1217246610.187596: WEXT: Operstate: linkmode=-1, operstate=5
1217246610.187607: EAPOL: External notification - portEnabled=0
1217246610.187615: EAPOL: External notification - portValid=0
1217246610.187619: EAPOL: External notification - EAP success=0
1217246610.220221: pairwise: 0x18
1217246610.220238: ssid - hexdump_ascii(len=4):
~ 65 64 72 32 edr2 1217246610.220248: group: 0x1e 
1217246610.220254: proto: 0x3
1217246610.220258: key_mgmt: 0x2
1217246610.220272: PSK - hexdump(len=32): [REMOVED]
1217246610.220277: scan_ssid=1 (0x1)
1217246610.220671: Setting scan request: 0 sec 0 usec
1217246610.220700: State: DISCONNECTED -> SCANNING
1217246610.220727: Starting AP scan (broadcast SSID)
1217246610.220742: Scan requested (ret=0) - scan timeout 30 seconds


Is there _any_ supplicant log output between these two log lines?  If
not, then the driver apparently refuses to scan or there aren't actually
any scan results.  But since you know there's an AP around, I'd suspect
driver problems here.  But the supplicant code in driver_wext.c looks
pretty fishy too since it doesn't handle the case where the driver
doesn't report any events, but finished the scan already.
There is absolutely nothing between these two lines. I've checked that on two laptops with Intel 4965, and they act exactly the same (both using kernel 2.6.26). One had 8 wireless network 
in range and the second one only one of them.
But this was always there. I remember having the same issue a year ago (when iwl4965 was in it's very early alpha stage). 
It seems FIRST scan takes long time and 2nd is instant.
Maybe driver takes longer than 30 sec to perform first scan? I have no idea at all how much time should it take to perform such scan, 
but isn't >4sec surprisingly quick?
Anyway, could you point me out the place, which needs to be modified to extend NM associate timeout? That way I could create my own patch and solve my problem. Although it might be worth taking under 
I'd really rather not, precisely because then the driver will continue
to suck.  If you're not getting authenticated within 30 - 45 seconds,
there is something horribly wrong (or your password is wrong or your AP
isn't actually there), and we should fix the problem at the source
rather than working around broken drivers.

Dan
I absolutely agree with you on this one. You might be able to fix the driver bug, but someone like me - simply can't. That's why I until driver works 100% as expected I prefer a simple workaround 
to avoid the frustration.
I've found the timeout I was looking for, changed it, tested over 50 reconnections and it asked me only once to retype my password (with default timeout it would ask at least once every 10 tries). 

How long did you have to make it, out of curiousity?

Dan
I made it 45 sec. Didn't test it with different settings yet, but I presume even 35 would do. 

How long should wireless scan take if the driver is good, out of curiosity?


Rob


Thanks for your help Dan. Really appreciate!

Rob

consideration to extend NM timeout by few sec. It shouldn't affect people
who got association in "standard" - below 30sec time - and can only help people who don't (correct me if I'm wrong here). 


Rob


Dan

1217246635.189546: No keys have been configured - skip key clearing
1217246635.189570: State: SCANNING -> DISCONNECTED
1217246635.189648: wpa_driver_wext_set_operstate: operstate 0->0 (DORMANT)
1217246635.189663: WEXT: Operstate: linkmode=-1, operstate=5
1217246635.189688: EAPOL: External notification - portEnabled=0
1217246635.189702: EAPOL: External notification - portValid=0
1217246635.189712: EAPOL: External notification - EAP success=0
1217246635.907342: Setting scan request: 0 sec 0 usec
1217246635.907377: State: DISCONNECTED -> SCANNING
1217246635.907406: Starting AP scan (broadcast SSID)
1217246635.907421: Scan requested (ret=0) - scan timeout 30 seconds

SECOND SCAN - 4 sec
1217246639.659847: RTM_NEWLINK: operstate=0 ifi_flags=0x1003 ([UP])
1217246639.659865: Wireless event: cmd=0x8b19 len=16
1217246639.659891: Received 298 bytes of scan results (1 BSSes)
1217246639.659898: CTRL-EVENT-SCAN-RESULTS

I'm _guessing_ that NM waits 30 sec for a association. If the first scan
timeouts it's got 5 sec for 2nd try. If supplicant won't make it
we've got:

NetworkManager: <info>  Activation (wlan0/wireless): association took
too long.

That would explain why sometimes it connects and sometimes it wont.


Rob

|
|>> Supplicant state 2 is "scanning" I think and indicates that the
|>> supplicant cannot find an compatible AP to associate with. Some causes 
|>> of that are invalid configuration (which probably isn't your problem
|>> since it does connect later) and odd drivers. What hardware and kernel 
|>> version are you using?
|> Network controller: Intel Corporation Device 4229 (rev 61)       <-
Intel 4965
|>
|>>> NetworkManager: <info>  wlan0: Device is fully-supported using driver
|>>> 'iwl4965'.
|>>> NetworkManager: <info>  wlan0: driver supports SSID scans (scan_capa
|> kernel - 2.6.26
|>
|>
|> Rob
|>
|>
|>> Dan
|>>
|>>> Here are some logs:
|>>>
|>>> 1) I start /usr/sbin/NetworkManager --no-daemon and
|>>> /etc/NetworkManager/system-setting directory is empty:
|>>>
|>>> networkmanager starts:
|>>> NetworkManager: <info>  Found radio killswitch
|>>> /org/freedesktop/Hal/devices/ipw_wlan_switch
|>>> NetworkManager: <info>  eth0: Device is fully-supported using driver
|>>> 'r8169'. NetworkManager: <info>  Found new Ethernet device 'eth0'.
|>>> NetworkManager: <info>  (eth0): exported as
|>>> /org/freedesktop/Hal/devices/net_00_1b_fc_4a_4a_20
|>>> NetworkManager: <info>  wlan0: Device is fully-supported using driver
|>>> 'iwl4965'.
|>>> NetworkManager: <info>  wlan0: driver supports SSID scans (scan_capa
|>>> 0x01). NetworkManager: <info>  Found new wireless (802.11) device
|>>> 'wlan0'. NetworkManager: <info>  (wlan0): exported as
|>>> /org/freedesktop/Hal/devices/net_00_13_e8_15_64_05
|>>> NetworkManager: <WARN>  killswitch_getpower_reply(): Error getting
|>>> killswitch power: hal-ipw-killswitch-linux returned 255.
|>>> NetworkManager: <info>  (eth0): device state change: 1 -> 2
|>>> NetworkManager: <info>  (eth0): bringing up device.
|>>> NetworkManager: <info>  (eth0): preparing device.
|>>> NetworkManager: <info>  (eth0): deactivating device.
|>>> NetworkManager: <info>  (wlan0): device state change: 1 -> 2
|>>> NetworkManager: <info>  (wlan0): bringing up device.
|>>> NetworkManager: <info>  (wlan0): preparing device.
|>>> NetworkManager: <info>  (wlan0): deactivating device.
|>>> NetworkManager: <info>  (wlan0): device state change: 2 -> 3
|>>> NetworkManager: <info>  (wlan0): supplicant interface state change:
1 ->
|>>> 2.
|>>>
|>>>
|>>> now I copy my configuration file to system-settings folder.
|>>> nm-system-settings picks it up instantly:
|>>>
|>>>
|>>> NetworkManager: <info>  Activation (wlan0) starting connection
'Wireless
|>>> connection X'
|>>> NetworkManager: <info>  (wlan0): device state change: 3 -> 4
|>>> NetworkManager: <info>  Activation (wlan0) Stage 1 of 5 (Device
Prepare)
|>>> scheduled...
|>>> NetworkManager: <info>  Activation (wlan0) Stage 1 of 5 (Device
Prepare)
|>>> started...
|>>> NetworkManager: <info>  Activation (wlan0) Stage 2 of 5 (Device
|>>> Configure) scheduled...
|>>> NetworkManager: <info>  Activation (wlan0) Stage 1 of 5 (Device
Prepare)
|>>> complete.
|>>> NetworkManager: <info>  Activation (wlan0) Stage 2 of 5 (Device
|>>> Configure) starting...
|>>> NetworkManager: <info>  (wlan0): device state change: 4 -> 5
|>>> NetworkManager: <info>  Activation (wlan0/wireless): connection
'Wireless
|>>> connection X' has security, and secrets exist. No new secrets needed. 
|>>> etworkManager: <info>  Config: added 'ssid' value 'dagger'
|>>> NetworkManager: <info>  Config: added 'scan_ssid' value '1'
|>>> NetworkManager: <info>  Config: added 'key_mgmt' value 'WPA-PSK'
|>>> NetworkManager: <info>  Config: added 'psk' value '<omitted>'
|>>> NetworkManager: <info>  Config: added 'pairwise' value 'TKIP CCMP'
|>>> NetworkManager: <info>  Config: added 'group' value 'TKIP CCMP'
|>>> NetworkManager: <info>  Activation (wlan0) Stage 2 of 5 (Device
|>>> Configure) complete.
|>>> NetworkManager: <info>  Config: set interface ap_scan to 1
|>>> NetworkManager: <info>  (wlan0): supplicant connection state
change: 1 ->
|>>> 2 NetworkManager: <info>  (wlan0): supplicant connection state
change: 2
|>>> -> 3 NetworkManager: <info>  (wlan0): supplicant connection state
change:
|>>> 3 -> 5 NetworkManager: <info>  (wlan0): supplicant connection state
|>>> change: 5 -> 4 NetworkManager: <info>  (wlan0): supplicant connection
|>>> state change: 4 -> 5 NetworkManager: <info>  (wlan0): supplicant
|>>> connection state change: 5 -> 6 NetworkManager: <info>  (wlan0):
|>>> supplicant connection state change: 6 -> 7 NetworkManager: <info>
|>>> Activation (wlan0/wireless) Stage 2 of 5 (Device Configure) successful. 
|>>> Connected to wireless network 'dagger'.
|>>> NetworkManager: <info>  Activation (wlan0) Stage 3 of 5 (IP Configure
|>>> Start) scheduled.
|>>> NetworkManager: <info>  Activation (wlan0) Stage 3 of 5 (IP Configure
|>>> Start) started...
|>>> NetworkManager: <info>  (wlan0): device state change: 5 -> 7
|>>> NetworkManager: <info> Activation (wlan0) Beginning DHCP transaction. 
|>>> Internet Systems Consortium DHCP Client V3.1.1-Gentoo
|>>> Copyright 2004-2008 Internet Systems Consortium.
|>>> All rights reserved.
|>>> For info, please visit http://www.isc.org/sw/dhcp/
|>>>
|>>> wmaster0: unknown hardware address type 801
|>>> NetworkManager: <info>  dhclient started with pid 23885
|>>> NetworkManager: <info>  Activation (wlan0) Stage 3 of 5 (IP Configure
|>>> Start) complete.
|>>>
|>>> everything as expected.
|>>>
|>>> 2) I start /usr/sbin/NetworkManager --no-daemon and
|>>> /etc/NetworkManager/system-setting directory contains my config file
|>>> we've got 2 cases (they are pretty random):
|>>>
|>>> case 1)
|>>>
|>>> etworkManager: <info>  Found radio killswitch
|>>> /org/freedesktop/Hal/devices/ipw_wlan_switch
|>>> NetworkManager: <info>  eth0: Device is fully-supported using driver
|>>> 'r8169'. NetworkManager: <info>  Found new Ethernet device 'eth0'.
|>>> NetworkManager: <info>  (eth0): exported as
|>>> /org/freedesktop/Hal/devices/net_00_1b_fc_4a_4a_20
|>>> NetworkManager: <info>  wlan0: Device is fully-supported using driver
|>>> 'iwl4965'.
|>>> NetworkManager: <info>  wlan0: driver supports SSID scans (scan_capa
|>>> 0x01). NetworkManager: <info>  Found new wireless (802.11) device
|>>> 'wlan0'. NetworkManager: <info>  (wlan0): exported as
|>>> /org/freedesktop/Hal/devices/net_00_13_e8_15_64_05
|>>> NetworkManager: <info>  Trying to start the system settings daemon...
|>>> NetworkManager: <WARN>  killswitch_getpower_reply(): Error getting
|>>> killswitch power: hal-ipw-killswitch-linux returned 255.
|>>> NetworkManager: <info>  (eth0): device state change: 1 -> 2
|>>> NetworkManager: <info>  (eth0): bringing up device.
|>>> NetworkManager: <info>  (eth0): preparing device.
|>>> NetworkManager: <info>  (eth0): deactivating device.
|>>> NetworkManager: <info>  (wlan0): device state change: 1 -> 2
|>>> NetworkManager: <info>  (wlan0): bringing up device.
|>>> NetworkManager: <info>  (wlan0): preparing device.
|>>> NetworkManager: <info>  (wlan0): deactivating device.
|>>> NetworkManager: <info>  (wlan0): device state change: 2 -> 3
|>>> NetworkManager: <info>  (wlan0): supplicant interface state change:
1 ->
|>>> 2. NetworkManager: <info>  Activation (wlan0) starting connection
|>>> 'Wireless connection X'
|>>> NetworkManager: <info>  (wlan0): device state change: 3 -> 4
|>>> NetworkManager: <info>  Activation (wlan0) Stage 1 of 5 (Device
Prepare)
|>>> scheduled...
|>>> NetworkManager: <info>  Activation (wlan0) Stage 1 of 5 (Device
Prepare)
|>>> started...
|>>> NetworkManager: <info>  Activation (wlan0) Stage 2 of 5 (Device
|>>> Configure) scheduled...
|>>> NetworkManager: <info>  Activation (wlan0) Stage 1 of 5 (Device
Prepare)
|>>> complete.
|>>> NetworkManager: <info>  Activation (wlan0) Stage 2 of 5 (Device
|>>> Configure) starting...
|>>> NetworkManager: <info>  (wlan0): device state change: 4 -> 5
|>>> NetworkManager: <info>  Activation (wlan0/wireless): connection
'Wireless
|>>> connection X' has security, and secrets exist. No new secrets needed. 
|>>> etworkManager: <info>  Config: added 'ssid' value 'dagger'
|>>> NetworkManager: <info>  Config: added 'scan_ssid' value '1'
|>>> NetworkManager: <info>  Config: added 'key_mgmt' value 'WPA-PSK'
|>>> NetworkManager: <info>  Config: added 'psk' value '<omitted>'
|>>> NetworkManager: <info>  Config: added 'pairwise' value 'TKIP CCMP'
|>>> NetworkManager: <info>  Config: added 'group' value 'TKIP CCMP'
|>>> NetworkManager: <info>  Activation (wlan0) Stage 2 of 5 (Device
|>>> Configure) complete.
|>>> NetworkManager: <info>  Config: set interface ap_scan to 1
|>>> NetworkManager: <info>  (wlan0): supplicant connection state
change: 2 ->
|>>> 0 NetworkManager: <info>  (wlan0): supplicant connection state
change: 0
|>>> -> 2
|>>>
|>>> NetworkManager: <info> wlan0: link timed out. <- TIMEOUT 
|>>> HERE
|>>>
|>>> NetworkManager: <info>  Activation (wlan0/wireless): association
took too
|>>> long.
|>>> NetworkManager: <info>  (wlan0): device state change: 5 -> 6
|>>> NetworkManager: <info>  Activation (wlan0/wireless): asking for new
|>>> secrets NetworkManager: <info>  (wlan0): supplicant connection state
|>>> change: 2 -> 0 NetworkManager: <WARN>  get_secrets_cb(): Couldn't get
|>>> connection secrets: nm- settings.c.307 - Missing implementation for
|>>> ConnectionSettings::get_secrets.. NetworkManager: <info>  (wlan0):
device
|>>> state change: 6 -> 9
|>>> NetworkManager: <info>  Activation (wlan0) failed for access point
|>>> (dagger) NetworkManager: <info>  Marking connection 'Wireless
connection
|>>> X' invalid. NetworkManager: <info>  Activation (wlan0) failed.
|>>> NetworkManager: <info>  (wlan0): device state change: 9 -> 3
|>>> NetworkManager: <info>  (wlan0): deactivating device.
|>>>
|>>> case 2)
|>>>
|>>> NetworkManager: <info>  Config: set interface ap_scan to 1
|>>> NetworkManager: <info>  (wlan0): supplicant connection state
change: 2 ->
|>>> 0 NetworkManager: <info>  (wlan0): supplicant connection state
change: 0
|>>> -> 2
|>>>
|>>> NetworkManager: <info>  wlan0: link timed out.               <- ALSO
|>>> TIMEOUT
|>>>
|>>> NetworkManager: <info>  (wlan0): supplicant connection state
change: 2 ->
|>>> 3 NetworkManager: <info>  (wlan0): supplicant connection state
change: 3
|>>> -> 5 NetworkManager: <info>  (wlan0): supplicant connection state
change:
|>>> 5 -> 4 NetworkManager: <info>  (wlan0): supplicant connection state
|>>> change: 4 -> 5 NetworkManager: <info>  (wlan0): supplicant connection
|>>> state change: 5 -> 6 NetworkManager: <info>  (wlan0): supplicant
|>>> connection state change: 6 -> 7 NetworkManager: <info>  Activation
|>>> (wlan0/wireless) Stage 2 of 5 (Device Configure) successful. Connected 
|>>> to wireless network 'dagger'.
|>>> NetworkManager: <info>  Activation (wlan0) Stage 3 of 5 (IP Configure
|>>> Start) scheduled.
|>>> NetworkManager: <info>  Activation (wlan0) Stage 3 of 5 (IP Configure
|>>> Start) started...
|>>>
|>>>
|>>> and it start successfully
|>>>
|>>>
|>>> In both cases it seems we always have supplicant state change:
|>>>
|>>> NetworkManager: <info>  (wlan0): supplicant connection state
change: 2 ->
|>>> 0 NetworkManager: <info>  (wlan0): supplicant connection state
change: 0
|>>> -> 2
|>>>
|>>> which doesn't occur if we copy config file later. Maybe NM should
wait a
|>>> bit longer before timing out.
|>>>
|>>> Also looking at case 1 NM tries to associate with AP only once and if
|>>> unsuccessful asks for password again. Maybe it would be worth changing 
|>>> this behavior to try associate at least twice before asking (please
note
|>>> it's not system-settings specific. It also happens with nm-applet if
|>>> first association wasn't successful - doesn't happen too often, but
it's
|>>> very annoying if you have to retype your password knowing you've
got the
|>>> right one in your keyring - even more annoying when your password
is 20+
|>>> characters long).
|>>>
|>>>
|>>> Cheers,
|>>> Rob
|>>>
|>>> On Sunday 27 July 2008 21:12:13 Robert Piasek wrote:
|>>>> Hi Dan,
|>>>>
|>>>>
|>>>> Thank you very much for the instructions and your help. I've tried to 
|>>>> do it myself using keyfile plugin specification, but I stacked on
SSID.
|>>>> I couldn't figure out how to specify that. I've spend few hours trying 
|>>>> before I gave up and posted to the mailing list. It would be good to
|>>>> stick it somewhere. (like
|>>>> http://live.gnome.org/NetworkManagerConfigurationSpecification or
|>>>> http://www.gnome.org/projects/NetworkManager/)
|>>>>
|>>>>
|>>>> Cheres,
|>>>> Rob
|>>>>
|>>>> On Sunday 27 July 2008 19:45:58 Dan Williams wrote:
|>>>>> On Sat, 2008-07-26 at 20:04 +0100, Robert Piasek wrote:
|>>>>>> Hi,
|>>>>>>
|>>>>>> Could anyone be kind enough to send me an example of keyfile plugin 
|>>>>>> file for wireless config?
|>>>>> [802-11-wireless-security]
|>>>>> key-mgmt=none
|>>>>> wep-tx-keyidx=0
|>>>>> auth-alg=open
|>>>>> wep-key0=33b7af2400ba80f060f144e195
|>>>>>
|>>>>> [connection]
|>>>>> id=Wireless connection 1
|>>>>> type=802-11-wireless
|>>>>> autoconnect=true
|>>>>> timestamp=0
|>>>>>
|>>>>> [802-11-wireless]
|>>>>> ssid=102;111;111;98;97;114;
|>>>>> mode=infrastructure
|>>>>> channel=0
|>>>>> rate=0
|>>>>> tx-power=0
|>>>>> mtu=0
|>>>>> security=802-11-wireless-security
|>>>>>
|>>>>> [ipv4]
|>>>>> method=manual
|>>>>> addresses1=192.168.1.2;24;192.168.1.1;
|>>>>> routes1=192.168.1.2;24;192.168.1.1;
|>>>>> ignore-dhcp-dns=false
|>>>>>
|>>>>> How I got this:
|>>>>>
|>>>>> 1) stop NetworkManager
|>>>>> 2) killall -TERM nm-system-settings
|>>>>> 3) /usr/sbin/nm-system-settings --debug --plugins=keyfile
|>>>>> 4) /usr/bin/nm-connection-editor
|>>>>> 5) Make a new wireless connection, fill everything out, then hit OK
|>>>>> 6) re-open the connection and check the "System connection" checkbox, 
|>>>>> then hit OK again
|>>>>>
|>>>>> Steps 5 & 6 may not need to be separate, that's just how I did them
|>>>>> because I hit OK too fast in step 5 then remembered why I was there
|>>>>> :)
|>>>>>
|>>>>> The routes1 field is also bogus here and shouldn't be filled in, but 
|>>>>> apparently we're screwing something up in the connection editor.
|>>>>>
|>>>>> Dan
|>>>>>
|>>>>>> On Saturday 19 July 2008 01:52:42 Robert Piasek wrote:
|>>>>>>> Hi,
|>>>>>>>
|>>>>>>> I would like to use system wide configuration for NM, but there
|>>>>>>> is no native plugin for my distribution. That leaves me with
|>>>>>>> keyfile plugin. I've tried to use it, but so far without any
|>>>>>>> luck. First I tried to find a decent example of configuration
|>>>>>>> file, but I could only find a specification page without any
|>>>>>>> examples. My tries to create a config file based on on that
|>>>>>>> specification failed.
|>>>>>>>
|>>>>>>> Than I tried to save my current config by selecting system
|>>>>>>> settings checkbox. But every time I click OK, it always comes up
|>>>>>>> with the error message "Adding connection failed: Launch helper
|>>>>>>> exited with unknown return code 1." (not very useful error
|>>>>>>> message :p).
|>>>>>> BTW this error message is generated when you try to save your
|>>>>>> connection as system-settings, and
|>>>>>> /etc/NetworkManager/nm-system-settings.conf does not exists.
|>>>>>>
|>>>>>>> Since system wide
|>>>>>>> settings are using policykit, so I'm _guessing_ it might be
|>>>>>>> related. I've added section below to my policykit.conf, but it
|>>>>>>> didn't change anything.
|>>>>>>>
|>>>>>>>
|>>>>>>> my PolicyKit.conf contains:
|>>>>>>>
|>>>>>>> <?xml version="1.0" encoding="UTF-8"?> <!-- -*- XML -*- -->
|>>>>>>>
|>>>>>>> <!DOCTYPE pkconfig PUBLIC "-//freedesktop//DTD PolicyKit
|>>>>>>> Configuration 1.0//EN"
|>>>>>>> "http://hal.freedesktop.org/releases/PolicyKit/1.0/config.dtd";>
|>>>>>>>
|>>>>>>> <!-- See the manual page PolicyKit.conf(5) for file format -->
|>>>>>>>
|>>>>>>> <config version="0.1">
|>>>>>>>      <match user="my_username">
|>>>>>>>          <return result="yes"/>
|>>>>>>>      </match>
|>>>>>>> </config>
|>>>>>>>
|>>>>>>>
|>>>>>>>
|>>>>>>> Any help appreciated.
|>>>>>>>
|>>>>>>>
|>>>>>>>
|>>>>>>> Cheers,
|>>>>>>> Rob
|>>>>>>>
|>>>>>>> _______________________________________________
|>>>>>>> NetworkManager-list mailing list
|>>>>>>> NetworkManager-list gnome org
|>>>>>>> http://mail.gnome.org/mailman/listinfo/networkmanager-list
|>>>>>> _______________________________________________
|>>>>>> NetworkManager-list mailing list
|>>>>>> NetworkManager-list gnome org
|>>>>>> http://mail.gnome.org/mailman/listinfo/networkmanager-list
|>>>> _______________________________________________
|>>>> NetworkManager-list mailing list
|>>>> NetworkManager-list gnome org
|>>>> http://mail.gnome.org/mailman/listinfo/networkmanager-list
|>> _______________________________________________
|>> NetworkManager-list mailing list
|>> NetworkManager-list gnome org
|>> http://mail.gnome.org/mailman/listinfo/networkmanager-list
|



_______________________________________________
NetworkManager-list mailing list
NetworkManager-list gnome org
http://mail.gnome.org/mailman/listinfo/networkmanager-list

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

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