--On Mittwoch, Dezember 27, 2006 21:36:46 +0100 Matthias Ferdinand 
<mf+tech-net@mfedv.net> wrote:
> To be really sure I had not messed with config or compiler or such, I
> downloaded the binary NetBSD 3.1 kernel (-GENERIC and -GENERIC_LAPTOP)
> from ftp.netbsd.org and tried again with both, but the result is exactly
> the same.

Instead of wiping my entire installation to try Linux, I did a cvs-upgrade 
for
/usr/src to "netbsd-4" and ran build.sh. Now with a 4.0 kernel:

bash-3.1# uname -a
NetBSD shlab 4.0_BETA2 NetBSD 4.0_BETA2 (GENERIC) #0: Thu Jan 11 09:38:18 
CET 2007  adminmf@shlab:/usr/obj/sys/arch/i386/compile/GENERIC i386

But I get the same result: pppd receives hangup immediately, no LCP packets
exchanged.

Then I tried with an additional USB<->serial adapter in between:
  PC->USB->serial_converter->RS232->mobile

this is an old USB 1.0 chipset:
    uhci0 at pci0 dev 7 function 2: Intel 82371AB (PIIX4) USB Host 
Controller (rev. 0x01)
    uhci0: interrupting at irq 11
    usb0 at uhci0: USB revision 1.0
    uhub0 at usb0
    uhub0: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
    uhub0: 2 ports with 2 removable, self powered

and this is the dmesg entry for the converter:
    uplcom0 at uhub0 port 1
    uplcom0: Prolific Technology Inc. USB-Serial Controller, rev 1.10/3.00, 
addr 2
    ucom0 at uplcom0

With ttyU0/dtyU0 instead of tty00/dty00 I get nearly the same results. 
What's different:
 a) the phone does not hangup after pppd has exited,
    only when the remote server closes the connection (about 1min later).
 b) kernel says "ppp0: got 24 bytes" instead of "ppp0: bad protocol 2206"
    Jan 15 01:20:38 shlab pppd[1049]: Serial connection established.
    Jan 15 01:20:38 shlab pppd[1049]: Using interface ppp0
    Jan 15 01:20:38 shlab pppd[1049]: Connect: ppp0 <--> /dev/dtyU0
    Jan 15 01:20:38 shlab pppd[1049]: Modem hangup
    Jan 15 01:20:38 shlab pppd[1049]: Connection terminated.
    Jan 15 01:20:38 shlab /netbsd: ppp0: got 24 bytes
    Jan 15 01:20:38 shlab /netbsd: 
ff03c0210101001402060000000005064e3dfd3107020802
    Jan 15 01:20:40 shlab pppd[1049]: Exit.
 c) a later chat invocation sees lots of noise, probably buffered ppp
    packets from previous connect:
    Jan 15 01:59:11 shlab chat[1307]: expect (OK)
    Jan 15 01:59:11 shlab chat[1307]: 
g&^H^F&^F^F&&^F^F^F^F^F&C^N&&F&lwg&^H^F&^F^F&&^F^F^F^F^FveFDEEFEng&^H^F^F^F^F&&^F^F^F^F^F&C^N&&F&Dwg&^H^F^F^F^F&&^F^F
    Jan 15 01:59:11 shlab chat[1307]: 
^F^F^F&C^N&&F&fwg&^H^F&^F^F&&^F^F^F^F^F&C^N&&F&-wg&^H^F&^F^F&vEDDDDDuC^N&&F&^Owg&^H^FF^F^F&&^F^F^F^F^F&C^N&&F&
    Jan 15 01:59:11 shlab chat[1307]: 
gwg&^H^FF^F^F&&^F^F^F^F^F&C^N&&F&Ewg&^H^Ff^F^F&&^F^F^F^F^F&C^N&&F&^Nwg&^H^Ff^FvDEEDDDDDeFDEEF^E&wg&^H^FF^F^F&

b and c may result from buffering in the serial adapter or from USB 
transferring blocks of data.


I think by now I have ruled out as source of the problem
 - the notebook rs232 interface (tty00)
     ttyU0 behaves almost the same
 - the data cable
     works reliably on a different PC
 - the phone / phone model
     works on a different PC (using Linux)
     three phones (Siemens ME45, 2x Siemens S35) behave exactly the same

This would leave the following suspects
 - kernel rs232/tty driver or settings
 - pppd tty initialisation

Are there perhaps some stty tricks I could try, which would not be 
overwritten by pppd? Other things I should look at?

Best regards
Matthias