asterisk users - Nov 2004 - snom200 -> asterisk & dtmf (rfc2833)

Hello all.

Seems, snom200 with 3.x software versions isn't
compliant to rfc2833, while no such problem with
2.04.g.
Timestamp and event duration are increased
simultaneously, while timestamp should points to the
beginning of the event and event duration extends
forwards from that time.

In any case, asterisk is able to detect dtmfs
correctly. But I believe that in scenarios where
asterisk acts as IP->PSTN gateway the dtmf sequence
generated by asterisk in PSTN side can be incorrect.
In other words dtmfs can be overlapped if they quickly
pressed on snom.

Unfortunately, I have no PSTN card installed on my
asterisk. 
Can somebody check if this works?

snom200(2.04.g vs 3.x)   --ip-->  asterisk  --pstn--> 
 phone

Thanks in advance,
Arsen.


		
__________________________________
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail

On Monday 01 November 2004 05:20 am, Arsen Chaloyan
wrote:
> Hello all.
>
> Seems, snom200 with 3.x software versions isn't
> compliant to rfc2833, while no such problem with
> 2.04.g.
> Timestamp and event duration are increased
> simultaneously, while timestamp should points to the
> beginning of the event and event duration extends
> forwards from that time.
>
> In any case, asterisk is able to detect dtmfs
> correctly. But I believe that in scenarios where
> asterisk acts as IP->PSTN gateway the dtmf sequence
> generated by asterisk in PSTN side can be incorrect.
> In other words dtmfs can be overlapped if they quickly
> pressed on snom.
>
> Unfortunately, I have no PSTN card installed on my
> asterisk.
> Can somebody check if this works?
>
> snom200(2.04.g vs 3.x)   --ip-->  asterisk  --pstn-->
>  phone
Hmm, I don't see that as even being an issue as snom only sends the digits 
once you press OK (except when asterisk is configured to read it
"live").

If you mean dtmf after a connection is established, then that's true for any
equipment. I.e. a too brief signal is always possible. I'd call that user 
error. Using my snom I can't say I've noticed anything like the above
during
normal use. 

Now there is another problem in that sometimes I don't get it to dial out. I
get a period of silence after hitting OK. A while later I get a dialtone. Or 
I end up with a fast busy. 

But I _think_ that is related to the non-digium hw I'm testing for pstn 
access. If I press redial a number of times I'll get through. My thought is 
that if it was a snom error then redial would never work, as it's only 
repeating what I typed. And, it behaves the same way if I manually redial a 
number of times.

My firmware is V3.52. 
-- 

Steve Szmidt

"They that would give up essential liberty for temporary safety 
deserve neither liberty nor safety."
                                Benjamin Franklin

Hello,
see my comments below.
 
On Monday 01 November 2004 05:20 am, Arsen Chaloyan
wrote:
> Hello all.
>
> Seems, snom200 with 3.x software versions isn't
> compliant to rfc2833, while no such problem with
> 2.04.g.
> Timestamp and event duration are increased
> simultaneously, while timestamp should points to the
> beginning of the event and event duration extends
> forwards from that time.
>
> Can somebody check if this works?
>
> snom200(2.04.g vs 3.x)   --ip-->  asterisk 
--pstn-->
>  phone
Hmm, I don't see that as even being an issue as snom
only sends the 
digits 
once you press OK (except when asterisk is configured
to read it 
"live"). 

If you mean dtmf after a connection is established,
then that's true 
for any 
equipment. I.e. a too brief signal is always possible.
I'd call that user error. Using my snom I can't say
I've noticed anything like the above 
during 
normal use. 

[Arsen] 
Thanks for response, Steve.
I just want to check the sequence of inband dtmfs
generated by asterisk (pstn leg) as a reply to
out-of-band dtmfs received from snom (ip leg).

Generally, I'm pretty happy with my snom phone and
with all changes which has been done since 2.04.g to
the 3.56, but I still cannot understand the change
regarding this technical issue with rtp stream and 
out-of-band dtmfs.
I try to dig into sources, which I have downloaded
from snom a month or two ago and seems find the change
in
snomaud.c

/* timestamp */
// buffer[4] = (unsigned char)((dtmf_start >> 24) &
0xff);
// buffer[5] = (unsigned char)((dtmf_start >> 16) &
0xff);
// buffer[6] = (unsigned char)((dtmf_start >> 8) &
0xff);
// buffer[7] = (unsigned char)(dtmf_start & 0xff);
buffer[4] = (unsigned char)((send_time >> 24) & 0xff);
buffer[5] = (unsigned char)((send_time >> 16) & 0xff);
buffer[6] = (unsigned char)((send_time >> 8) & 0xff);
buffer[7] = (unsigned char)(send_time & 0xff); 

Strange, but the right part is commented there.
send_time is used instead of dtmf_start.

I know, asterisk has nothing to do with this, but can
anybody please give some comments on this?

Best regards, 
Arsen.

P.S. Hope, this isn't spam for you.


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com