more on 0x3302 Protocol error layer 2

more on 0x3302 Protocol error layer 2

Post by microte » Thu, 22 Jan 2004 19:25:26

Dear Tobias,

my CAPI msg formatter shows ALL bytes belonging to the MSG (based on 1
L0022 I0001 M8002 U0002 C01 P00 N0000
- 10 00 0B 80 30 34 33 32 35 37 30 39 30 31 00 00 00 00 00 00 00 00

In the particular case of CONNECT_REQ, follows an expanded view:


L 0022 full msg len: 34 bytes long
I 0001 applic ID
M 8002 CONN_REQ cmd ID
U 0002 msg ID
C 01 controller
N 0000 NCCI
10 00 CIP value: 16 Telephony -> BC:(0x04,0x03,0x80,0x90,0xA3)
HLC: (0x7D, 0x02,0x91,0x81)
0B Called Party Number: 11 bytes
80 Type of number and numbering plan
30 34 33 32 35 37 30 39 30 31 Digits

00 Calling Party Number
00 Called Party Subaddress
00 Calling Party Subaddress
00 B protocol: default => B1prot: 64kbps with HDLC
B2prot: ISO7776 (X.75 SLP)
B3prot: Transparent
B1conf: <empty>
B2conf: Link Address A: 0x03;
Link Address B: 0x01;
Modulo mode:8; Window
B3conf: <empty>
GlobalConf: B Channel Op
(DTE/DCE): Default
00 BC (default from CIP setting)
00 LLC (default from CIP setting)
00 HLC (default from CIP setting)
00 Additional Info: default => B channel identific.
Information: choose ANY

One more hint:
Digi's connection tester, when used to make an outbound call,
gives the same sequence and the same error my programs report.
(To be fair, in their instructions they yet say to use it only
to make a "CALL RING" to destination and not a "full CALL").

Thanks to you and everyone interesting in this puzzle...


more on 0x3302 Protocol error layer 2

Post by Jochen Kle » Fri, 23 Jan 2004 17:57:17


AFAIK from your first thread you want to establish a *voice* connection,
isnt it?
Or do you want to establish a HDLC/X.75 based 64 KBps data link, as shown in
the traces youve provided now?!??
Sorry I am bit confused by the different infromation in both of your
threads, now.

Jochen Klein

< XXXX@XXXXX.COM > schrieb im Newsbeitrag


more on 0x3302 Protocol error layer 2

Post by microte » Wed, 28 Jan 2004 00:03:14

In the first "thread" ('0x3302 error protocol layer 2'),
I put the whole sequence of CAPI commands and indications
(and their respective "byte sequences") which produced the error.

In order to clarify (for Tobias) the setup of the outbound call,
I took from that sequence the single CONNECT_REQ command
(with NO modifications) and I expanded and commented
the contents of all parameters in a subsequent "thread".
('more on 0x3302 error protocol layer 2', this thread).

I simply detailed the defaults of CAPI's CONNECT_REQ fields:

B protocol is 0, which means "default protocol behavior:
ISO 7776 and windows size seven (Layer 1:HDLC framing protocol,
Layer 2:X.75 SLP and Layer 3: transparent)

Bearer Capabilities is 0, and so LLC:0 and HLC:0
accepting the parameters resulting from CIP selection

My first choice about CIP parameter was: speech [1],
which did not work.
After an analysis of the succesful outbound calls made by Phoner
and similar applications, I saw they use CIP parameter:
Telephony [16] and so I decided to use it (yet with no results).

Summing up, I'd like to make "the simplest voice outbound call"...
and that's the reason why I put every possible struct defaults in

Bye bye

more on 0x3302 Protocol error layer 2

Post by Tobias Eri » Wed, 28 Jan 2004 15:52:54

Hi again,

you won't get any connection (layer3) established with a voice-terminal
on the other side, if you don't specify bittransparent operation. What
you selected will try to establish a real layer3 data-connection, which
will not start up because the other side will not respond to the HDLC
sequences that you side sends.


< XXXX@XXXXX.COM > schrieb im Newsbeitrag