<p>Sometimes faxing works without any configuration, which can be misleading. The difference between passthrough and straight g.711 is a longer jitter buffer, disabling dtmf, and disabling the echo canceller. </p>
<p>That is odd that a zte device would send proprietary cisco nse packets. Maybe that is what changed, the gateway?</p>
<p>I don't think they are using pt 96 for dtmf because the event numbers generally match the dtmf digit pressed.</p>
<p>Instead of seeing nse packets you should be seeing a sip reinvite with a t.38 sdp since it should ve a protocol based negotiation.</p>
<div class="gmail_quote">On Jan 17, 2011 12:22 PM, "Antonio Soares" <<a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a>> wrote:<br type="attribution">> It’ s a SIP trunk between a Cisco GW (the one I have control) and a ZTE<br>
> device… It was working with the config I have shown. Something was changed<br>> on the other side. I was trying to make it work using the options available<br>> at our side. Unfortunately I don’t have the debugs captured when this was<br>
> configured so I still have a doubt. What PT should be used in this T.38<br>> negotiation ? PT=100 or PT=101 ?<br>> <br>> <br>> <br>> PT=100: Designates an NSE (Named Signaling Event packet). NSE packets can<br>
> signal a variety of different messages so it is helpful to know the NSE<br>> Event ID.<br>> <br>> <br>> <br>> PT=101: Designates an NTE (Named Telephony Event) as defined in RFC 2833.<br>> <br>> <br>
> <br>> Check this table:<br>> <br>> <br>> <br>> <a href="https://supportforums.cisco.com/docs/DOC-1387">https://supportforums.cisco.com/docs/DOC-1387</a><br>> <br>> <br>> <br>> I was convinced that it was PT=100 (NSE) but as you said, they are Cisco<br>
> proprietary. So maybe it should be PT=101 (NTE) ?<br>> <br>> <br>> <br>> <br>> <br>> Thanks.<br>> <br>> <br>> <br>> Regards,<br>> <br>> <br>> <br>> Antonio Soares, CCIE #18473 (R&S/SP)<br>
> <mailto:<a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a>> <a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a><br>> <br>> <br>> <br>> From: <a href="mailto:matthn@gmail.com">matthn@gmail.com</a> [mailto:<a href="mailto:matthn@gmail.com">matthn@gmail.com</a>] On Behalf Of Nick Matthews<br>
> Sent: segunda-feira, 17 de Janeiro de 2011 17:56<br>> To: Antonio Soares<br>> Cc: Peter Slow; <a href="mailto:cisco-voip@puck.nether.net">cisco-voip@puck.nether.net</a><br>> Subject: Re: [cisco-voip] T.38 Fax Issue<br>
> <br>> <br>> <br>> When you say SIP trunk, are you implying that this is going to a SIP<br>> provider? Or is this just a SIP gateway and the fax is going out a TDM<br>> interface?<br>> <br>> If it's a SIP trunk you're sending the fax over, forget about anything using<br>
> NSEs - like Pete said they're Cisco proprietary.<br>> <br>> You'll need to configure the other device for protocol based T.38 switchover<br>> like you've got it configured on your end.<br>> <br>
> -nick<br>> <br>> On Mon, Jan 17, 2011 at 10:21 AM, Antonio Soares <<a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a>><br>> wrote:<br>> <br>> We tried the "fax protocol t38 nse force" but it didn't work.<br>
> <br>> Now with "fallback cisco" we have something different:<br>> <br>> --------------------------------------<br>> s=VoIP d=DSP payload 0x60 ssrc 0xFFD5D4C sequence 0x291 timestamp<br>
> 0x19A50<br>> <br>> <<<Rcv> Pt:96 Evt:136 Pkt:01 40 50<br>> --------------------------------------<br>> <br>> s=DSP d=VoIP payload 0x61 ssrc 0xFFE2 sequence 0x0 timestamp 0x0<br>
> Pt:97 Evt:0 Pkt:00 04 00 <Snd>>><br>> --------------------------------------<br>> s=DSP d=VoIP payload 0x60 ssrc 0x0 sequence 0x0 timestamp 0x0<br>> Pt:96 Evt:0 Pkt:00 00 00 <Snd>>><br>
> --------------------------------------<br>> s=VoIP d=DSP payload 0x60 ssrc 0xFFD5D4C sequence 0x292 timestamp<br>> 0x19AA0<br>> <br>> <<<Rcv> Pt:96 Evt:136 Pkt:01 40 50<br>> --------------------------------------<br>
> <br>> s=DSP d=VoIP payload 0x61 ssrc 0x60014 sequence 0x1 timestamp 0x1<br>> Pt:97 Evt:0 Pkt:01 00 00 <Snd>>><br>> --------------------------------------<br>> s=VoIP d=DSP payload 0x60 ssrc 0xFFD5D4C sequence 0x293 timestamp<br>
> 0x19AF0<br>> <br>> <<<Rcv> Pt:96 Evt:136 Pkt:01 40 50<br>> --------------------------------------<br>> <br>> s=DSP d=VoIP payload 0x61 ssrc 0x0 sequence 0x2 timestamp 0x2<br>
> Pt:97 Evt:0 Pkt:01 00 00 <Snd>>><br>> --------------------------------------<br>> s=VoIP d=DSP payload 0x60 ssrc 0xFFD5D4C sequence 0x294 timestamp<br>> 0x19B40<br>> <br>
> <<<Rcv> Pt:96 Evt:136 Pkt:01 40 50<br>> --------------------------------------<br>> <br>> s=DSP d=VoIP payload 0x61 ssrc 0x0 sequence 0x3 timestamp 0x3<br>> Pt:97 Evt:0 Pkt:01 00 00 <Snd>>><br>
> --------------------------------------<br>> s=DSP d=VoIP payload 0x7A ssrc 0xFFFFFFFF sequence 0x9EBF<br>> timestamp 0x2FE4CED0<br>> Pt:122 Evt:0 Pkt:52 3A 00 <Snd>>><br>
> --------------------------------------<br>> <br>> Now we see:<br>> <br>> Pt:96 Evt:136 (Sent by a voice gateway to signal a switchover to Cisco<br>> fax relay)<br>> Pt:97 Evt:0 (Sent by a voice gateway to confirm a successful switchover<br>
> to Cisco fax relay)<br>> Pt:122 Evt:0 (Indicates Cisco fax relay data)<br>> <br>> So I was expecting this would mean it was working but apparently it still<br>> doesn't work.<br>> <br>> I'm using this document as reference:<br>
> <br>> Cisco RTP Payload Types<br>> <br>> <a href="https://supportforums.cisco.com/docs/DOC-1387">https://supportforums.cisco.com/docs/DOC-1387</a><br>> <br>> Any other ideas ?<br>> <br>> <br>> <br>
> Thanks.<br>> <br>> Regards,<br>> <br>> Antonio Soares, CCIE #18473 (R&S/SP)<br>> <a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a><br>> <br>> <br>> -----Original Message-----<br>
> From: Peter Slow [mailto:<a href="mailto:peter.slow@gmail.com">peter.slow@gmail.com</a>]<br>> <br>> Sent: segunda-feira, 17 de Janeiro de 2011 12:16<br>> To: Antonio Soares<br>> Cc: <a href="mailto:cisco-voip@puck.nether.net">cisco-voip@puck.nether.net</a><br>
> Subject: Re: [cisco-voip] T.38 Fax Issue<br>> <br>> No, "fallback cisco" is different. You need to try the "fax protocol<br>> t38 nse force" command.<br>> <br>> -pete<br>> <br>> <br>
> On Mon, Jan 17, 2011 at 6:58 AM, Antonio Soares <<a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a>> wrote:<br>>> It was working before so I assume the other end's configuration was<br>> changed.<br>
>> This is a SIP trunk and we have this configuration for the T.38 fax:<br>>><br>>> voice service voip<br>>> fax protocol t38 ls-redundancy 0 hs-redundancy 0 fallback pass-through<br>>> g711alaw<br>
>><br>>> We are now trying the following:<br>>><br>>> voice service voip<br>>> fax protocol t38 ls-redundancy 0 hs-redundancy 0 fallback cisco<br>>><br>>> This should match their PT=96 request, right ?<br>
>><br>>><br>>> Thanks.<br>>><br>>> Regards,<br>>><br>>> Antonio Soares, CCIE #18473 (R&S/SP)<br>>> <a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a><br>>><br>
>><br>>> -----Original Message-----<br>>> From: Peter Slow [mailto:<a href="mailto:peter.slow@gmail.com">peter.slow@gmail.com</a>]<br>>> Sent: segunda-feira, 17 de Janeiro de 2011 11:50<br>>> To: Antonio Soares<br>
>> Cc: <a href="mailto:cisco-voip@puck.nether.net">cisco-voip@puck.nether.net</a><br>>> Subject: Re: [cisco-voip] T.38 Fax Issue<br>>><br>>> Usually t.38 Fax relay operates by switching to t.38 based on an MGCP<br>
>> MDCX, a new h245 TCS/OLC etc, or a SIP re-invite. this is referred to<br>>> as protocol based switchover, and is generally how people's gateways<br>>> are configured for t.38 when it is being used. it would appear that<br>
>> the other GW you are talking to is set up for NSE based switchover,<br>>> which AFAIK is a cisco proprietary method. You may be configured for<br>>> protocol based switchover still, while it sounds like perhaps the<br>
>> other side went and changed something and is now trying to do NSE<br>>> based switchover...<br>>><br>>> Check out this link:<br>>><br>> <a href="http://www.cisco.com/en/US/docs/ios/12_3/vvf_c/cisco_ios_fax_services_over_i">http://www.cisco.com/en/US/docs/ios/12_3/vvf_c/cisco_ios_fax_services_over_i</a><br>
>> p_application_guide/t38.html#wp1156699<br>>><br>>> It is tough to help you out beyond what I've said without knowing what<br>>> protocol (h.323, MGCP..) you are using for call control. More insight<br>
>> there, plus perhaps some asn1 or mgcp packet debugs would help, but<br>>> try setting your side to use nse based switchover if it isnt set that<br>>> way already.<br>>><br>>> -Peter<br>>><br>
>><br>>><br>>> On Mon, Jan 17, 2011 at 5:48 AM, Antonio Soares <<a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a>><br>> wrote:<br>>>> Hello group,<br>>>><br>>>> A few days ago, fax stopped working on one gateway. The "debug voip rtp<br>
>>> session nse" gives me this:<br>>>><br>>>><br>>><br>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++<br>>>> ++++++++++++++<br>>>> s=VoIP d=DSP payload 0x60 ssrc 0x509B391F sequence 0x18E<br>
>> timestamp<br>>>> 0xF8C0<br>>>> <<<Rcv> Pt:96 Evt:136 Pkt:02 80 50<br>>>> s=VoIP d=DSP payload 0x60 ssrc 0x509B391F sequence 0x18F<br>>> timestamp<br>
>>> 0xF910<br>>>> <<<Rcv> Pt:96 Evt:136 Pkt:01 40 50<br>>>> s=VoIP d=DSP payload 0x60 ssrc 0x509B391F sequence 0x190<br>>> timestamp<br>>>> 0xF960<br>
>>> <<<Rcv> Pt:96 Evt:136 Pkt:01 40 50<br>>>> s=VoIP d=DSP payload 0x60 ssrc 0x509B391F sequence 0x191<br>>> timestamp<br>>>> 0xF9B0<br>>>> <<<Rcv> Pt:96 Evt:136 Pkt:01 40 50<br>
>>> s=VoIP d=DSP payload 0x60 ssrc 0x509B391F sequence 0x192<br>>> timestamp<br>>>> 0xFA00<br>>>> (...)<br>>>><br>>><br>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++<br>
>>> ++++++++++++++<br>>>><br>>>> Most likely, the problem is on the other end, a gateway that I do not<br>>>> control.<br>>>><br>>>> I believe for this to work I need to receive PT=100 and EVT=200:<br>
>>><br>>>> Payload Type 100: Designates an NSE (Named Signaling Event packet). NSE<br>>>> packets can signal a variety of different messages so it is helpful to<br>>> know<br>>>> the NSE Event ID.<br>
>>><br>>>> Event 200: Triggered by the detection of a V.21 fax preamble when T.38 is<br>>>> configured. This message initiates the switchover of the call from voice<br>>> to<br>>>> T.38. This message is sent by the terminating fax gateway to notify the<br>
>>> originating fax gateway of the switch to T.38.<br>>>><br>>>> And I'm receiving PT=96 and EVT=136. PT=96 means " Sent by a voice<br>> gateway<br>>>> to signal a switchover to Cisco fax relay". I wasn't able to find the<br>
>>> meaning of EVT=136.<br>>>><br>>>> With this information, can I assume that the problem is on the other end<br>> ?<br>>>> Can I do anything on my side to force the desired PT/EVT combination ?<br>
>>><br>>>><br>>>> Thanks.<br>>>><br>>>> Regards,<br>>>><br>>>> Antonio Soares, CCIE #18473 (R&S/SP)<br>>>> <a href="mailto:amsoares@netcabo.pt">amsoares@netcabo.pt</a><br>
>>><br>>>><br>>>> _______________________________________________<br>>>> cisco-voip mailing list<br>>>> <a href="mailto:cisco-voip@puck.nether.net">cisco-voip@puck.nether.net</a><br>
>>> <a href="https://puck.nether.net/mailman/listinfo/cisco-voip">https://puck.nether.net/mailman/listinfo/cisco-voip</a><br>>>><br>>><br>>><br>> <br>> <br>> _______________________________________________<br>
> cisco-voip mailing list<br>> <a href="mailto:cisco-voip@puck.nether.net">cisco-voip@puck.nether.net</a><br>> <a href="https://puck.nether.net/mailman/listinfo/cisco-voip">https://puck.nether.net/mailman/listinfo/cisco-voip</a><br>
> <br>> <br>> <br></div>