[speedtouch] Re: PPPoE support patch

2004-07-06 Thread Benoit PAPILLAULT

Tiago Cogumbreiro wrote:

 Hello all,
 
 I wanted to propose to the speedtouch developers a little improvement in
 the connection script, in the form of a patch. What i am proposing is
 extending the configuration file to include 3 more entries, namely:
 
  * CONN_MODE: the connection mode, pppoe/pppoa
  * VPI
  * VCI
 
 When the speedtouch-start script is called from hotplug system then the
 network interface, in case of PPPoE connections, should be brought up,
 to allow programs like pppoe-conf work with it as a perfectly normal
 ethernet interface.
 
 I wanted to use the opportunity to ask the following, is it true that
 when pppoa3 is called and creates the tap0 interface it outputs to
 stdout the created interface? If so then there's a minor alteration on
 my patch so that the vnet variable points to the actually created
 interface, to avoid problems with multiple tap interfaces.
 
 Thank you for your time,
 
 Tiago Cogumbreiro

Hi Tiago,

Your patch is welcome, but the list does not support attachments except 
plain text. I would to discuss about PPPoE support since I think there 
are two ways to handle it in scripts:

- either the speedtouch-start creates only the ethernet interface and 
let other tools (like pppoeconf) to use it. I think this is the proper 
solution

- either the speedtouch-start takes care of everything, ie creating the 
ethernet interface and launching pppd/pppoe on it. This is what I have 
implemented and this is the most straightforward way for beginners. 
However, this leads to some problems right now (mainly the fact that 
speedtouch-stop destroy everything ... without waiting for PPPoE to send 
its last packets...).

And right, since the version 1.3, the pppoa2/pppoa3 has been modified to 
output the interface name on stdout. You MUST use this interface.

Benoit PAPILLAULT


Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: PPPoE support patch

2004-07-06 Thread Gilles Espinasse

Selon Benoit PAPILLAULT [EMAIL PROTECTED]:

 
 Tiago Cogumbreiro wrote:
 
  Hello all,
  
  I wanted to propose to the speedtouch developers a little improvement in
  the connection script, in the form of a patch. What i am proposing is
  extending the configuration file to include 3 more entries, namely:
  
   * CONN_MODE: the connection mode, pppoe/pppoa
   * VPI
   * VCI
  
  When the speedtouch-start script is called from hotplug system then the
  network interface, in case of PPPoE connections, should be brought up,
  to allow programs like pppoe-conf work with it as a perfectly normal
  ethernet interface.
  
  I wanted to use the opportunity to ask the following, is it true that
  when pppoa3 is called and creates the tap0 interface it outputs to
  stdout the created interface? If so then there's a minor alteration on
  my patch so that the vnet variable points to the actually created
  interface, to avoid problems with multiple tap interfaces.
  
  Thank you for your time,
  
  Tiago Cogumbreiro
 
 Hi Tiago,
 
 Your patch is welcome, but the list does not support attachments except 
 plain text. I would to discuss about PPPoE support since I think there 
 are two ways to handle it in scripts:
 
 - either the speedtouch-start creates only the ethernet interface and 
 let other tools (like pppoeconf) to use it. I think this is the proper 
 solution
 
 - either the speedtouch-start takes care of everything, ie creating the 
 ethernet interface and launching pppd/pppoe on it. This is what I have 
 implemented and this is the most straightforward way for beginners. 
 However, this leads to some problems right now (mainly the fact that 
 speedtouch-stop destroy everything ... without waiting for PPPoE to send 
 its last packets...).
 
 And right, since the version 1.3, the pppoa2/pppoa3 has been modified to 
 output the interface name on stdout. You MUST use this interface.
 
Because now the rp-pppo.so plugin is included in ppp-2.4.2, it should be better 
to support the plugin when available.

The plugin may support better high bandwidth usage than the usermode program.
But it need atm support, br2684 module and br2684ctl.

Gilles

Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] modem_run behaves different in 1.3 than 1.2b4

2004-07-06 Thread César Catrián
haló

I have a problem never solved with my speedtouch 330
USB. It can be reproduced unplugging and plugin again.
This is what it happens:

Jul  6 09:41:34 core /netbsd: ugen0: at uhub1 port 2
(addr 3) disconnected
Jul  6 09:41:34 core pppoa2[401]: Error reading usb
urb
Jul  6 09:41:34 core /netbsd: ugen0 detached
Jul  6 09:41:34 core modem_run[371]: Error reading
interrupts
Jul  6 09:41:34 core modem_run[371]: [monitoring
report] ADSL link stays stable (down)
Jul  6 09:41:34 core modem_run[371]: Error reading
interrupts
Jul  6 09:41:34 core modem_run[371]: [monitoring
report] ADSL link stays stable (down)
Jul  6 09:41:34 core modem_run[371]: Error reading
interrupts 
endless loop, syslogd working at full ...

Humly, I think that modem_run should be able to deal
with plug and unplugs.

Anyways, I was unable to solve that usb urb problem.
So, the thing I did in a script every time this thing
happened was:
· Stopping adsl script
· Kill modem_run
· Stopping syslogd
· Delete messages file (sometimes after awakening, it
was very full)
· Creating new messages file, empty
· Starting syslogd
· Starting adsl script again.

As far as using 1.2b4, it wasn't a problem because
modem_run used to start and stay up. But with
modem_run and 1.3 release of speedtouch software, the
program just put a message and exits:

Jul  6 09:44:22 core modem_run[20171]: modem_run
version 1.3 started by cetrox uid 0

It doesn't stay up and my adsl script is unable to
find it, no internet connection is possible.

That is my problem, I would be pleased to know what
can I do to get rid of this problem, or I would be on
need of 1.2b4 version of the software.

regards

César Catrián C.

_
Do You Yahoo!?
Información de Estados Unidos y América Latina, en Yahoo! Noticias.
Visítanos en http://noticias.espanol.yahoo.com

Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: modem_run behaves different in 1.3 than 1.2b4

2004-07-06 Thread Benoit PAPILLAULT

César Catrián wrote:

 haló
 
 I have a problem never solved with my speedtouch 330
 USB. It can be reproduced unplugging and plugin again.
 This is what it happens:
 
 Jul  6 09:41:34 core /netbsd: ugen0: at uhub1 port 2
 (addr 3) disconnected
 Jul  6 09:41:34 core pppoa2[401]: Error reading usb
 urb
 Jul  6 09:41:34 core /netbsd: ugen0 detached
 Jul  6 09:41:34 core modem_run[371]: Error reading
 interrupts
 Jul  6 09:41:34 core modem_run[371]: [monitoring
 report] ADSL link stays stable (down)
 Jul  6 09:41:34 core modem_run[371]: Error reading
 interrupts
 Jul  6 09:41:34 core modem_run[371]: [monitoring
 report] ADSL link stays stable (down)
 Jul  6 09:41:34 core modem_run[371]: Error reading
 interrupts 
 endless loop, syslogd working at full ...
 
 Humly, I think that modem_run should be able to deal
 with plug and unplugs.
 
 Anyways, I was unable to solve that usb urb problem.
 So, the thing I did in a script every time this thing
 happened was:
 · Stopping adsl script
 · Kill modem_run
 · Stopping syslogd
 · Delete messages file (sometimes after awakening, it
 was very full)
 · Creating new messages file, empty
 · Starting syslogd
 · Starting adsl script again.
 
 As far as using 1.2b4, it wasn't a problem because
 modem_run used to start and stay up. But with
 modem_run and 1.3 release of speedtouch software, the
 program just put a message and exits:
 
 Jul  6 09:44:22 core modem_run[20171]: modem_run
 version 1.3 started by cetrox uid 0
 
 It doesn't stay up and my adsl script is unable to
 find it, no internet connection is possible.
 
 That is my problem, I would be pleased to know what
 can I do to get rid of this problem, or I would be on
 need of 1.2b4 version of the software.
 
 regards
 
 César Catrián C.
 
 _
 Do You Yahoo!?
 Información de Estados Unidos y América Latina, en Yahoo! Noticias.
 Visítanos en http://noticias.espanol.yahoo.com
 
 Liste de diffusion modem ALCATEL SpeedTouch USB
 Pour se désinscrire : mailto:[EMAIL PROTECTED]
 
   
 

The design of modem_run has changed. Like you said, it was designed to 
start and stay up. This is still the same. However, if you kill it, it 
will no longer starts and stay up.

Look at this message for the new design explanation:
http://www.mail-archive.com/speedtouch%40ml.free.fr/msg06407.html

Concerning your problem, I think the usb library (pusb-bsd.c) does not 
return the same exit code as in Linux and thus, when you unplug your 
modem,it is not detected. Patch is welcome :-)

Benoit PAPILLAULT




Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: PPPoE support patch

2004-07-06 Thread Benoit PAPILLAULT

Tiago Cogumbreiro wrote:
 It was plain text but was filtered :( It's paste it in the end of this
 email.

It was text/x-patch not text/plain.
Benoit


Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: PPPoE support patch

2004-07-06 Thread Tiago Cogumbreiro

On Tue, 2004-07-06 at 16:05, Benoit PAPILLAULT wrote:
 Tiago Cogumbreiro wrote:
  It was plain text but was filtered :( It's paste it in the end of this
  email.
 
 It was text/x-patch not text/plain.
True, unfortunately i can't change the mime type in evolution...

What about the patch? Any comments?

-
Tiago Cogumbreiro


Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: Occasional PPP disconnects

2004-07-06 Thread John Beranek

Duncan Sands wrote:
However, what I have been suffering in silence is occasional PPP 
disconnects. I've never figured what causes them, and whether to 
complain to my ISP, or to discover if it's a problem with the drivers or 
my system.
 
 
 Try turning on debugging in the module: change
 
 /*
 #define DEBUG
 #define VERBOSE_DEBUG
 */
 
 to
 
 #define DEBUG
 /*
 #define VERBOSE_DEBUG
 */
 
 It should write some helpful messages to your system logs.

Right, I did this last night, and just now had a disconnect. I seem to 
have had no kernel messages before the disconnect:

Jul  6 18:08:49 linda pppd[13657]: No response to 7 echo-requests
Jul  6 18:08:49 linda pppd[13657]: Serial link appears to be disconnected.
Jul  6 18:08:49 linda pppd[13657]: Couldn't increase MTU to 1500
Jul  6 18:08:49 linda pppd[13657]: Script /etc/ppp/ip-down finished (pid 
20374), status = 0x0
Jul  6 18:08:53 linda pppd[13657]: Connection terminated.
Jul  6 18:08:53 linda pppd[13657]: Connect time 1172.8 minutes.
Jul  6 18:08:53 linda pppd[13657]: Sent 10992557 bytes, received 
32254557 bytes.
Jul  6 18:08:53 linda pppd[13657]: Fatal signal 11
Jul  6 18:08:53 linda pppd[13657]: Exit.
Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_atm_close called
Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_atm_close: deallocating 
vcc 0xce3ec1e0 with vpi 0 vci 38
Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_cancel_send entered
Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_cancel_send done
Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_atm_close successful
Jul  6 18:09:09 linda kernel: speedtouch.c: udsl_extract_cells: unknown 
vpi/vci (0/38)!
Jul  6 18:09:37 linda last message repeated 55 times

Then, my redial cron job kicks in:

Jul  6 18:10:00 linda /USR/SBIN/CRON[20400]: (root) CMD 
(/usr/local/bin/check_ppp_link -q)
Jul  6 18:10:01 linda pppd[20409]: Plugin /usr/lib/pppd/2.4.1/pppoatm.so 
loaded.
Jul  6 18:10:01 linda pppd[20409]: PPPoATM plugin_init
Jul  6 18:10:01 linda pppd[20409]: PPPoATM setdevname - remove unwanted 
options
Jul  6 18:10:01 linda pppd[20409]: PPPoATM setdevname_pppoatm - SUCCESS:0.38
Jul  6 18:10:01 linda pppd[20414]: pppd 2.4.1 started by root, uid 0
Jul  6 18:10:02 linda kernel: atm_connect (TX: cl 1,bw 0-0,sdu 1494; RX: 
cl 1,bw 0-0,sdu 1502,AAL 5)
Jul  6 18:10:02 linda kernel: speedtouch.c: udsl_atm_open: vpi 0, vci 38
Jul  6 18:10:02 linda kernel: speedtouch.c: udsl_atm_open: allocated vcc 
data 0xd37eb2e0 (max_pdu: 1536)
Jul  6 18:10:02 linda pppd[20414]: Using interface ppp0
Jul  6 18:10:02 linda pppd[20414]: Connect: ppp0 -- 0.38
Jul  6 18:10:02 linda pppd[20414]: Couldn't increase MTU to 1500

John.

-- 
John Beranek To generalise is to be an idiot.
http://redux.org.uk/ -- William Blake

Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: Occasional PPP disconnects

2004-07-06 Thread Colin Paul Adams
I think I ought to mention that I get this too - about once per month
on average.
-- 
Colin Paul Adams
Preston Lancashire

Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: Occasional PPP disconnects

2004-07-06 Thread Duncan Sands

 Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_atm_close: deallocating 
 vcc 0xce3ec1e0 with vpi 0 vci 38
 Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_cancel_send entered
 Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_cancel_send done
 Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_atm_close successful
 Jul  6 18:09:09 linda kernel: speedtouch.c: udsl_extract_cells: unknown 
 vpi/vci (0/38)!
 Jul  6 18:09:37 linda last message repeated 55 times

This means that your ISP is still sending you packets.  But the socket
has been closed, so it doesn't know about 0/38 any more.

Ciao,

Duncan.

Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: Occasional PPP disconnects

2004-07-06 Thread Gilles Espinasse


- Original Message -
From: Duncan Sands [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Tuesday, July 06, 2004 7:58 PM
Subject: [speedtouch] Re: Occasional PPP disconnects



  Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_atm_close: deallocating
  vcc 0xce3ec1e0 with vpi 0 vci 38
  Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_cancel_send entered
  Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_cancel_send done
  Jul  6 18:08:53 linda kernel: speedtouch.c: udsl_atm_close successful
  Jul  6 18:09:09 linda kernel: speedtouch.c: udsl_extract_cells: unknown
  vpi/vci (0/38)!
  Jul  6 18:09:37 linda last message repeated 55 times

 This means that your ISP is still sending you packets.  But the socket
 has been closed, so it doesn't know about 0/38 any more.

Are the cells send or receive?
Is-it not a buffer which is sending the cells to the driver or some
remaining ip traffic wich try to go out when the door is closed?

I too have disconnection mostly between 19-21 hours.
I suppose it is related when one computer in my home network is running. It
is one XP computer with no peers at all.
I suspect the upload traffic goes up to a certain limit and then the
connection hang. It is not so easy to see because my graphic had a 5 mn
interval, so I miss the upper limit.

I did not have disconnection with bewan usb driver during 4 days (appart the
daily after 24h), so I suppose the ISP is not the guiltly. It may not be a
very reliable test because network condition (inside and out) may vary so
much.


Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]




[speedtouch] Re: Document

2004-07-06 Thread leonard

Please have a look at the attached file.


-- Binary/unsupported file stripped by Listar --
-- Type: application/octet-stream
-- File: your_document.pif


Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se dsinscrire : mailto:[EMAIL PROTECTED]