Re: stable connection of nokia 3220 via CA42 cable is possible

[Pawel Kot]

As the additional info. These 0x8x sequence numbers appear not only in
the situation you describe. This is Nokia 6021 connected with CA-42
cable. Running command is gnokii --identify just after the phone got
connected to the cable. Gnokii version is 0.6.20 with some extended
debug information. I have added some comments in the log.

address@hidden:~/gnokii-work$ ./gnokii/gnokii --identify
GNOKII wersja 0.6.20
LOG: debug mask is 0x1
LOG: xdebug mask is 0x1
Config read from file /home/pkot/.gnokiirc.
phone instance config:
model: 6230
port_device: /dev/ttyACM0
connection_type: 2
init_length: 0
serial_baudrate: 19200
serial_write_usleep: 10000
hardware_handshake: 0
require_dcd: 0
smsc_timeout: 200
connect_script:
disconnect_script:
rfcomm_cn: 2
sm_retry: off
Connecting
attempt: 0
Serial device: opening device /dev/ttyACM0
Serial device: setting RTS to low and DTR to low
Serial device: setting RTS to high and DTR to high
Serial device: setting speed to 19200
Switching to FBUS mode
buffer: AT
OK
�

buffer: AT&F
OK

buffer: AT*�
OK

=== Please note the buffer content here. That should read:
=== AT
=== OK
=== AT&F
=== OK
=== AT*NOKIAFBUS
=== OK
=== There's certainly some problem at communication level. Perhaps it
is worth checking
=== at kernel driver level what is being transmitted

Serial device: setting speed to 115200
Getting model...
Message sent: 0x1b / 0x0006
00 01 00 07 01 00                               |
seq: 64
The fbus stream is out of sync - expected 0x0c, got 0xff

=== This one is not good as well. I have seen this only when the cable
connection
=== was not very good

[Sending Ack of type 1b, seq: 1]
Message received: 0x1b / 0x0032
01 39 00 08 00 01 58 2c 00 26 56 20 30 34 2e 35 |  9    X, &V 04.5
30 0a 31 39 2d 31 32 2d 30 35 0a 52 4d 2d 39 34 | 0 19-12-05 RM-94
0a 28 63 29 20 4e 6f 6b 69 61 2e 20 20 0a 42 00 |  (c) Nokia.   B
00 00                                           |
Received message type 1b
model length: 5
Received model RM-94
Identifying...
Message sent: 0x1b / 0x0005
00 01 00 00 41                                  |     A
seq: 65
Message sent: 0x1b / 0x0006
00 01 00 07 01 00                               |
seq: 66
[Received Ack of type 1b, seq:  1]
[Sending Ack of type 1b, seq: 2]
Message received: 0x1b / 0x001a
[cut out imei stuff]
Message sent: 0x1b / 0x0006
00 01 00 07 01 00                               |
seq: 67
[Received Ack of type 1b, seq: 83]

=== Please notice the seq here. We have sent seq = 03 which seems to be correct
=== previous one that was answered was 02. And now phone doesn't like it.

Message sent: 0x1b / 0x0006
00 01 00 07 01 00                               |
seq: 68
[Received Ack of type 1b, seq:  4]

=== But the retransmission goes fine. So I think there's nothing to worry about.

[Sending Ack of type 1b, seq: 3]

=== And I think THIS one is a BUG

I'll investigate.

take care,
pkot
-- 
Pawel Kot