[asterisk-r2] Sending DTMF 'f' ?

Moises Silva moises.silva at gmail.com
Wed Mar 18 19:44:11 CDT 2009 

As I said before, the Sangoma or Digium drivers cannot magically pass
the DTMF to the Asterisk core, it has to first be requested and read
by the chan_dahdi.c Asterisk driver, if you want to really find the
culprit you need to provide the log I asked for.

Moy

On Wed, Mar 18, 2009 at 4:06 PM, Amilcar Silvestre <amilcar at vonix.com.br> wrote:
> Moises,
>
> 27 minutes working with no problems at all. No more "DTMF "f" received
> on DAHDI" at all. The problem is on wanpipe hardware dtmf detection.
>
> But this is very strange since, as i said, i use TDMV_HW_DTMF enabled
> on many, many clients without any problems.
>
> Since now you're working for Sangoma, do you have any clues why is
> this happening? ;-) I'm using wanpipe 3.3.16 (hmm, the other boxes are
> running 3.3.14, maybe the new version introduced some bug to the
> hardware dtmf detector?)
>
> Regards,
> Amilcar.
>
>
> On Mar 18, 2009, at 5:39 PM, Amilcar Silvestre wrote:
>
>> Moises,
>>
>> No actions from manager here.
>>
>> I'm suspecting that this comes from the hardware DTMF detection in the
>> echo cancelation module in the board (Sangoma A104D), altough i've
>> never seem this happened, and I have at lest 60 others boxes with R2
>> and hardware dtmf detection using sangoma. Works like a charm. I've
>> turned it off. Let us see if the problem persists.
>>
>> Regards,
>> Amilcar.
>>
>> On Mar 18, 2009, at 5:33 PM, Moises Silva wrote:
>>
>>> No is not enough to see DAHDI/3-1, because that message is printed by
>>> the core, but the core may have dequeued this DTMF digit from other
>>> source than a read() in chan_dahdi.c, as I said, it may have come
>>> from
>>> the manager (which is capable of enqueueing DTMF in any channel), and
>>> there is other ways to fake incoming DTMF.
>>>
>>> Moy
>>>
>>> On Wed, Mar 18, 2009 at 5:29 PM, Amilcar Silvestre <amilcar at vonix.com.br
>>>> wrote:
>>>> The "DTMF Down " message will only be seem if i turn DEBUG level to
>>>> the logs. DMTF only is not sufficient. Doing it now.
>>>>
>>>> Amilcar.
>>>>
>>>> On Mar 18, 2009, at 5:23 PM, Amilcar Silvestre wrote:
>>>>
>>>>> Moises,
>>>>>
>>>>> DTMF begin 'f'received on DAHDI/3-1 doesn't mean that the DTMF is
>>>>> coming from DAHDI??
>>>>>
>>>>> No DTMF Down anywhere in the logs.
>>>>>
>>>>> Amilcar.
>>>>>
>>>>> On Mar 18, 2009, at 5:15 PM, Moises Silva wrote:
>>>>>
>>>>>> I still fail to see the log where this DTMF comes from
>>>>>> chan_dahdi.c
>>>>>>
>>>>>> Did you just miss it?
>>>>>>
>>>>>> Can you check your log and see if you can find a message like:
>>>>>>
>>>>>> DTMF Down 'f'
>>>>>>
>>>>>> in chan_dahdi?? otherwise somehow that DTMF is received from
>>>>>> somewhere
>>>>>> else, perhaps a manager application or something like that?
>>>>>>
>>>>>> On Wed, Mar 18, 2009 at 4:51 PM, Amilcar Silvestre <amilcar at vonix.com.br
>>>>>>> wrote:
>>>>>>> Moises,
>>>>>>>
>>>>>>> The log from console, with dtmf debug enabled is:
>>>>>>>
>>>>>>> [Mar 18 17:49:59] DTMF[3411]: channel.c:2279 __ast_read: DTMF
>>>>>>> begin
>>>>>>> 'f' received on DAHDI/3-1
>>>>>>> [Mar 18 17:49:59] DTMF[3411]: channel.c:2289 __ast_read: DTMF
>>>>>>> begin
>>>>>>> passthrough 'f' on DAHDI/3-1
>>>>>>> [Mar 18 17:49:59] WARNING[3411]: rtp.c:2205
>>>>>>> ast_rtp_senddigit_begin:
>>>>>>> Don't know how to represent 'f'
>>>>>>>
>>>>>>> Amilcar.
>>>>>>>
>>>>>>> On Mar 18, 2009, at 4:47 PM, Amilcar Silvestre wrote:
>>>>>>>
>>>>>>>> Hi Moyses,
>>>>>>>>
>>>>>>>> Ok, I've enabled DTMF debugging, and here's what i've got:
>>>>>>>>
>>>>>>>> [Mar 18 17:34:12] NOTICE[2690] chan_dahdi.c: MFC/R2 call has
>>>>>>>> been
>>>>>>>> accepted on chan 10
>>>>>>>> [Mar 18 17:34:12] NOTICE[2690] chan_dahdi.c: Call accepted on
>>>>>>>> forward
>>>>>>>> channel 10
>>>>>>>> (...)
>>>>>>>> [Mar 18 17:39:51] DTMF[2705] channel.c: DTMF begin 'f' received
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:39:51] DTMF[2705] channel.c: DTMF begin passthrough
>>>>>>>> 'f'
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:39:51] WARNING[2705] rtp.c: Don't know how to
>>>>>>>> represent
>>>>>>>> 'f'
>>>>>>>> (...)
>>>>>>>> [Mar 18 17:40:17] DTMF[2705] channel.c: DTMF begin 'f' received
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:40:17] DTMF[2705] channel.c: DTMF begin passthrough
>>>>>>>> 'f'
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:40:17] WARNING[2705] rtp.c: Don't know how to
>>>>>>>> represent
>>>>>>>> 'f'
>>>>>>>> (...)
>>>>>>>> [Mar 18 17:40:24] DTMF[2705] channel.c: DTMF begin 'f' received
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:40:24] DTMF[2705] channel.c: DTMF begin passthrough
>>>>>>>> 'f'
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:40:24] WARNING[2705] rtp.c: Don't know how to
>>>>>>>> represent
>>>>>>>> 'f'
>>>>>>>> (...)
>>>>>>>> [Mar 18 17:40:44] DTMF[2705] channel.c: DTMF begin 'f' received
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:40:44] DTMF[2705] channel.c: DTMF begin passthrough
>>>>>>>> 'f'
>>>>>>>> on
>>>>>>>> DAHDI/10-1
>>>>>>>> [Mar 18 17:40:44] WARNING[2705] rtp.c: Don't know how to
>>>>>>>> represent
>>>>>>>> 'f'
>>>>>>>> [Mar 18 17:40:45] DTMF[2793] channel.c: DTMF begin 'f' received
>>>>>>>> on
>>>>>>>> DAHDI/12-1
>>>>>>>> [Mar 18 17:40:45] DTMF[2793] channel.c: DTMF begin passthrough
>>>>>>>> 'f'
>>>>>>>> on
>>>>>>>> DAHDI/12-1
>>>>>>>> [Mar 18 17:40:45] WARNING[2793] rtp.c: Don't know how to
>>>>>>>> represent
>>>>>>>> 'f'
>>>>>>>> (...)
>>>>>>>> [Mar 18 17:43:05] NOTICE[2705] chan_dahdi.c: Chan 10 - Far end
>>>>>>>> disconnected. Reason: Forced Release
>>>>>>>> [Mar 18 17:43:05] NOTICE[2705] chan_dahdi.c: MFC/R2 call
>>>>>>>> disconnected
>>>>>>>> on chan 10
>>>>>>>> [Mar 18 17:43:05] NOTICE[4775] chan_dahdi.c: MFC/R2 call end on
>>>>>>>> chan
>>>>>>>> 10
>>>>>>>>
>>>>>>>> And this is happening in two different telcos.
>>>>>>>>
>>>>>>>> Yes, I've patched asterisk. The patch is that from googlecode
>>>>>>>> for
>>>>>>>> 1.4.23, with a very tiny modification to make the channel
>>>>>>>> returns
>>>>>>>> the
>>>>>>>> hangupcause.
>>>>>>>>
>>>>>>>> Amilcar.
>>>>>>>>
>>>>>>>>
>>>>>>>> On Mar 18, 2009, at 4:32 PM, Moises Silva wrote:
>>>>>>>>
>>>>>>>>> Hum, I see what you mean. But, MFC and DTMF use different pair
>>>>>>>>> of
>>>>>>>>> frequencies and F does not even exist in DTMF. Please enable
>>>>>>>>> dtmf
>>>>>>>>> debugging in your logger.conf and try to reproduce, I'd expect
>>>>>>>>> to
>>>>>>>>> see
>>>>>>>>> clearly if chan_dahdi/chan_zap is the one detecting a MF digit
>>>>>>>>> and
>>>>>>>>> wrongly sending it down to the core as DTMF digit.
>>>>>>>>>
>>>>>>>>> Did you patched that asterisk yourself?
>>>>>>>>>
>>>>>>>>> From a quick look at the code, the 'f' frame subclass is also
>>>>>>>>> used
>>>>>>>>> for
>>>>>>>>> FAX, so that f does not necessarily refers to the MF F tone.
>>>>>>>>>
>>>>>>>>> Moy
>>>>>>>>>
>>>>>>>>> On Wed, Mar 18, 2009 at 4:20 PM, Amilcar Silvestre <amilcar at vonix.com.br
>>>>>>>>>> wrote:
>>>>>>>>>> Hi Moises,
>>>>>>>>>>
>>>>>>>>>> I've already recorded the call using mixmonitor on the sip
>>>>>>>>>> side.
>>>>>>>>>> The
>>>>>>>>>> same on the recorded file. Mixmonitor only records the sip
>>>>>>>>>> side,
>>>>>>>>>> and
>>>>>>>>>> the far end is muted.
>>>>>>>>>>
>>>>>>>>>> And the problem only occurs on the R2 links. PRI works ok
>>>>>>>>>> (same
>>>>>>>>>> box).
>>>>>>>>>>
>>>>>>>>>> Seems that, after it enters on the function
>>>>>>>>>> ast_rtp_senddigit_begin
>>>>>>>>>> in
>>>>>>>>>> rtp.c, the function returns 0 (it doesn't have the "f" digit
>>>>>>>>>> for
>>>>>>>>>> DTMF,
>>>>>>>>>> the digit 'f' is related to MFC). After it returns 0, the
>>>>>>>>>> audio
>>>>>>>>>> from
>>>>>>>>>> DAHDI stops been redirected to the sip end point.
>>>>>>>>>>
>>>>>>>>>> Amilcar.
>>>>>>>>>>
>>>>>>>>>> On Mar 18, 2009, at 4:08 PM, Moises Silva wrote:
>>>>>>>>>>
>>>>>>>>>>> I think you should try to post this in asterisk-users, this
>>>>>>>>>>> is
>>>>>>>>>>> not
>>>>>>>>>>> an R2 issue.
>>>>>>>>>>>
>>>>>>>>>>> Having said that, it would be a good idea to reproduce, and
>>>>>>>>>>> then,
>>>>>>>>>>> when
>>>>>>>>>>> you have a call like that, use dahdi_monitor or ztmonitor to
>>>>>>>>>>> verify
>>>>>>>>>>> the audio is getting into the board correctly, then you can
>>>>>>>>>>> monitor
>>>>>>>>>>> the RTP traffic on the SIP side of the call and see if
>>>>>>>>>>> Asterisk
>>>>>>>>>>> is
>>>>>>>>>>> still sending the audio correctly, if it does, then the
>>>>>>>>>>> problem
>>>>>>>>>>> is
>>>>>>>>>>> definitely not even in Asterisk.
>>>>>>>>>>>
>>>>>>>>>>> On Wed, Mar 18, 2009 at 3:54 PM, Amilcar Silvestre <amilcar at vonix.com.br
>>>>>>>>>>>> wrote:
>>>>>>>>>>>> Douglas,
>>>>>>>>>>>> If you meant the CAS bits, is 1101.
>>>>>>>>>>>> If you mean the configuration of the link:
>>>>>>>>>>>> signalling=mfcr2
>>>>>>>>>>>> mfcr2_variant=br
>>>>>>>>>>>> mfcr2_get_ani_first=no
>>>>>>>>>>>> mfcr2_max_ani=10
>>>>>>>>>>>> mfcr2_max_dnis=4
>>>>>>>>>>>> mfcr2_category=national_subscriber
>>>>>>>>>>>> mfcr2_logdir=span1
>>>>>>>>>>>> mfcr2_logging=all
>>>>>>>>>>>> mfcr2_metering_pulse_timeout=500
>>>>>>>>>>>> Regards,
>>>>>>>>>>>> Amilcar.
>>>>>>>>>>>>
>>>>>>>>>>>> On Mar 18, 2009, at 3:50 PM, Douglas Fischer wrote:
>>>>>>>>>>>>
>>>>>>>>>>>> What do you have in your start bits? (zapata.conf)
>>>>>>>>>>>>
>>>>>>>>>>>> 2009/3/18 Amilcar Silvestre <amilcar at vonix.com.br>
>>>>>>>>>>>>>
>>>>>>>>>>>>> Hi,
>>>>>>>>>>>>>
>>>>>>>>>>>>> I have a box using asterisk 1.4.23.2, and OpenR2 1.1.0. The
>>>>>>>>>>>>> board
>>>>>>>>>>>>> is a
>>>>>>>>>>>>> Sangoma A104D, From times to times, it shows this message:
>>>>>>>>>>>>>
>>>>>>>>>>>>> WARNING[31548] rtp.c: Don't know how to represent 'f'
>>>>>>>>>>>>>
>>>>>>>>>>>>> Seems to be something related to sending a DTMF digit.
>>>>>>>>>>>>> After
>>>>>>>>>>>>> this
>>>>>>>>>>>>> message, the caller (an internal SIP endpoint) doens't hear
>>>>>>>>>>>>> anything
>>>>>>>>>>>>> more from the called (far end, coming from a R2 link), but
>>>>>>>>>>>>> the
>>>>>>>>>>>>> far
>>>>>>>>>>>>> end
>>>>>>>>>>>>> keeps receiving the audio from the SIP end point.
>>>>>>>>>>>>>
>>>>>>>>>>>>> The message comes with no intervention from the user on SIP
>>>>>>>>>>>>> end
>>>>>>>>>>>>> point.
>>>>>>>>>>>>>
>>>>>>>>>>>>> Does anyone knows what is happening?
>>>>>>>>>>>>>
>>>>>>>>>>>>> Thanks,
>>>>>>>>>>>>> Amilcar.
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>>>>>>>> digital.com--
>>>>>>>>>>>>>
>>>>>>>>>>>>> asterisk-r2 mailing list
>>>>>>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> --
>>>>>>>>>>>> Douglas Fernando Fischer
>>>>>>>>>>>> Engº de Controle e Automação
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>>>>>>> digital.com--
>>>>>>>>>>>>
>>>>>>>>>>>> asterisk-r2 mailing list
>>>>>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>>>>>>
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>>>>>>> digital.com--
>>>>>>>>>>>>
>>>>>>>>>>>> asterisk-r2 mailing list
>>>>>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> --
>>>>>>>>>>> "I do not agree with what you have to say, but I’ll defend to
>>>>>>>>>>> the
>>>>>>>>>>> death your right to say it." Voltaire
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>>>>>> digital.com--
>>>>>>>>>>>
>>>>>>>>>>> asterisk-r2 mailing list
>>>>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>>>>> digital.com--
>>>>>>>>>>
>>>>>>>>>> asterisk-r2 mailing list
>>>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> "I do not agree with what you have to say, but I’ll defend to
>>>>>>>>> the
>>>>>>>>> death your right to say it." Voltaire
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>>>> digital.com--
>>>>>>>>>
>>>>>>>>> asterisk-r2 mailing list
>>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>>> digital.com--
>>>>>>>>
>>>>>>>> asterisk-r2 mailing list
>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>>> digital.com--
>>>>>>>
>>>>>>> asterisk-r2 mailing list
>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> "I do not agree with what you have to say, but I’ll defend to the
>>>>>> death your right to say it." Voltaire
>>>>>>
>>>>>> _______________________________________________
>>>>>> --Bandwidth and Colocation Provided by http://www.api-
>>>>>> digital.com--
>>>>>>
>>>>>> asterisk-r2 mailing list
>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>>>>>
>>>>> asterisk-r2 mailing list
>>>>> To UNSUBSCRIBE or update options visit:
>>>>>  http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>
>>>>
>>>> _______________________________________________
>>>> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>>>>
>>>> asterisk-r2 mailing list
>>>> To UNSUBSCRIBE or update options visit:
>>>>  http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>>
>>>
>>>
>>>
>>> --
>>> "I do not agree with what you have to say, but I’ll defend to the
>>> death your right to say it." Voltaire
>>>
>>> _______________________________________________
>>> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>>>
>>> asterisk-r2 mailing list
>>> To UNSUBSCRIBE or update options visit:
>>>  http://lists.digium.com/mailman/listinfo/asterisk-r2
>>
>>
>> _______________________________________________
>> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>>
>> asterisk-r2 mailing list
>> To UNSUBSCRIBE or update options visit:
>>   http://lists.digium.com/mailman/listinfo/asterisk-r2
>
>
> _______________________________________________
> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>
> asterisk-r2 mailing list
> To UNSUBSCRIBE or update options visit:
>   http://lists.digium.com/mailman/listinfo/asterisk-r2
>



-- 
"I do not agree with what you have to say, but I’ll defend to the
death your right to say it." Voltaire

More information about the asterisk-r2 mailing list