May 2012 - FreeSWITCH-users - FreeSWITCH Mailing Lists

Re: [Freeswitch-users] DTMF Passthrough on different legs
by JRichey 02 Jun '12

02 Jun '12

What CODECs are being used? In-band DTMF won't work correctly with compressed CODECs like G.729. -----Original Message----- From: freeswitch-users-bounces(a)lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org]On Behalf Of Michael Lutz Sent: Thursday, May 31, 2012 2:12 AM To: FreeSWITCH Users Help Subject: [Freeswitch-users] DTMF Passthrough on different legs Hi Guys, I am stuck with a bit of a problem I cannot figure out where it is going wrong.. Here's my situation, I have an inbound call, comming from an external provider, they don't support RFC2833 so I turn on inband detection using 'start_dtmf' in the dialplan. This works ok, for the incomming channel and IVR provided (via Lua) to the caller. (Though sometimes it seems to misinterpret and returns wrong keys..) (Session:A) After that, this call is bridged (via Lua to a a new inbound session on the same switch) (without specifying start_dtmf) so RC2388 is used by default (?) (Session:B) This new sessions is playing soem audio (Lua) and is heard by the original caller (the two sessions are bridged). Seperatly I have an inbound ESL connection where I do an origanate using a originate {execute_on_answer=start_dtmf, .... etc) &park() to setup an new external call via the same gateway (they don't support RFC2388 so I activate the inband detection on the b-leg to receive the digits. Let's call this leg Session:C. After the orignated call is answered I bridge the parked call with the previous session (Session:B), so at the end Session:A and Session:B can talk to each other. Al this works perfectly. [Incommming call] Session:A ----- INBAND ------> Bridge(FS.Lua) ---> Session:B ---- RFC2388 ----> FS.Bridge ---> Session:C --- INBAND ------> ESL destination number My symptoms are... 1. I have a customer who had a IVR on his own side, (Session:C) and requires input from Session:A. The weird thing is that it seems the input is wrong. Keys entered do not correspond to what his IVR is receiving. As if newly dtmfs are generated somewhere. When I switch to another inbound provider, making Session:A using rfc2388 and not inband, but do dialout via the provider using inband detection. It seems to work fine. 2. Not being able to get the dtmf's from the ESL Destination number (Session:C). I'm reading this by subscribing to events via ESL for that particulair uuid. When I have a connection (inbound) without inband detection it seems to get the DTMF's from Session:C without any problems. Which makes it weird for me. note: I know the setup is a bit odd, but Session A cannot exit it's Lua script, so I needed to do a workaround to be able to keep te lua alive and being able to control the call (audio) via ESL. I realy need some help on this one, becuase we have a lot of customers complaining about not responding to dtmf's or receving invalid input in the (FS) IVR. Thanks for you help, Regards, Mike. _________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting(a)freeswitch.org http://www.freeswitchsolutions.com FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com Join Us At ClueCon - Aug 7-9, 2012 FreeSWITCH-users mailing list FreeSWITCH-users(a)lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

3 3

NAT issues and "best practices"
by Sean Devoy 01 Jun '12

01 Jun '12

HI All, I have a customer location that has just been a nightmare to implement. I am just learning that they "may" have multiple NAT routers in sequence at their location. I think I fully understand what NAT is trying to accomplish. There seem to be different levels and approaches. The most basic NAT setup (to me) is a HOME LAN with multiple PCs where NAT allows multiple devices to share a single routable IP address on the WAN side from multiple local devices the LAN side. Note I said OUTBOUND initiated connections. Even FTP can have trouble with this level. Almost all inbound traffic is blocked for security. Clearly for FS we need the switch to be able to punch through from the WAN to specific local IPs on the LAN to reach specific phones. This is INBOUND NAT and brings up many security issues for people. Even on devices where you get this "working" you may only be able to support one line per phone or a single inbound connection at a time. I understand NAT has PMP and UPnP protocols and FS "supports" both. What I can't find is where someone says "Here is a great setup that works with cheap, available "commodity" hardware from Cisco/Linksys that supports all the NAT you need for FS." I don't care if it is PMP or UPnP and I might not even care why you pick one over the other, although it is probably a "good read". Can someone just stand up say "FS works GREAT with the XYZ router in ABC mode from MY COMPANY using NAT to Cisco phones"? I have seen some articles about Freeware/Shareware firmware in this devices, but as a novice I want to limit the unknowns until I get more up to speed. My specific issue now is that I cannot get SCA to work at the NAT location. I issued: sofia_contact 220(a)mydomain.com sofia/external/sip:200@<ip address>:44234,sofia/external/sip:200@<same ip>:1024 Only one phone rings on inbound and the line indicator light does not change when either is picked up. Same configuration is working on our LAN with the switch. I am absolutely ready to by a router to fix these issues, I don't want to lose this customer. Thanks for your thoughts, Sean

4 4

Custom Registration
by alberto Villa 01 Jun '12

01 Jun '12

Hello, I'm using FreeSWITCH in a very custom system and I need to setup this kind of registration: There are two SIP phones A and B: when each SIP phone have its own unique ID the registration to freeswitch goes as usual and no change is needed, but when A and B share the same ID I would like to reject the registration for the last SIP phone. So, if A and B have the same ID and A is registered first, I would like to reject the subsequent registration of B. I tried to change the values of some configuration parameters such as "sip-allow-multiple-registrations" and "max-registrations-per-extension" but it didn't work (by the way I'm not shure to have edited properly this parameters). How can I achive this kind of registration? Thank you Alberto -- Dr. Villa Alberto Sw Engineer SeleTech srl via Collodi, 8 20052 Monza (MI) tel: +39 039 5962000 fax: +39 039 9716905 email: a.villa(a)seletech.com web: www.seletech.com www.seletech.eu

3 3

FreeTDM CallerID Detection with DTMF
by Antonio Modesto 01 Jun '12

01 Jun '12

Hi, I have a Digium TDM410P card, I am using it here in Brazil, where the signaling is not FSK, it's DTMF. It is working, though I am not receiving the callerid: Caller-Caller-ID-Number: [0000000000] Here is my freetdm.conf: [span zt FXO1] fxo-channel => 1 [span zt FXO2] fxo-channel => 2 [span zt FXO3] fxo-channel => 3 Here is one section of my autoload_configs/freetdm.conf.xml: <span name="FXO1">    <param name="tonegroup" value="br"/> <param name="digit-timeout" value="2000"/> <param name="max-digits" value="11"/> <param name="dialplan" value="XML"/> <param name="context" value="public"/> <param name="enable-callerid" value="true"/>     </span> Is it possible to enable the callerid detection in these conditions, or is it a hardware/driver limitation? Regards.

3 4

FSClient a Windows FreeSWITCH softphone Update Released
by Mitch Capper 01 Jun '12

01 Jun '12

An update had been posted for FSClient is a windows .NET softphone that uses libfreeswitch at the core. Info page at: http://wiki.freeswitch.org/wiki/FSClient Direct binary download: http://files.freeswitch.org/windows/installer/x86/FSClient.zip And source is in contrib repo under MitchCapper/FSClient Some of the changes include: Built against FS 1.2 RC isac codec support for contacts you can now right click and choose what account to call from if sofia goes offline update the accounts status Jabra headset improvements to avoid some crash conditions background task improvements to avoid the phone breaking without you knowing Feedback always welcome. With ZRTP into core now the next version may have some ZRTP related features. ~Mitch

4 3

Re: [Freeswitch-users] Problem with missing / duplicated DTMF
by Michael Lutz 31 May '12

31 May '12

Hi, Picking up on an older thread, I was wondering if this was ever resolved and how, or if there are more people experiencing the same issues (now or in the past) I seem to be experiencing the same issue! A lot of the times DTMF's are not recognized (and do not show up in events or debug log) and apart from I have noticed recognition of wrong dtmfs too. Sometimes I even see 'recognized' dtmfs while no key is pressed at al! (of which I am sure because it was my own test call) Hope someone is able to help me with this! I was looking in the combination of inband detection, because I use that for most of my calls. However reading this thread it appears that at least someone experienced same behaviour without enabling inband detection using start_dtmf. Regards, Mike. >Anthony Minessale anthony.minessale at gmail.com >Tue Feb 1 21:22:45 MSK 2011 >I dont think there any current dtmf issues open. >It sounds like maybe you are going across the pstn and encountering >some problems with transition from >2833 to inband and back again or from hair-pinning the call. > On Tue, Feb 1, 2011 at 12:00 PM, Matt Stockton <mstockton at harqen.com> wrote: > Sorry about not including the version. The version of freeswitch I am using > is. > FreeSWITCH Version 1.0.head (git-256a82d 2011-01-31 10-12-28 -0600) > I just updated to the latest yesterday to re-test it. > On Tue, Feb 1, 2011 at 10:10 AM, Matt Stockton <mstockton at harqen.com> wrote: >> >> I have having trouble with both missing and duplicated DTMF in >> Freeswitch. >> Here are the steps of how I am using it: >> 1. Leg A - I am calling out from my Freeswitch instance (through iCall), >> and I am calling an iCall number that is also connected to the same >> Freeswitch instance. >> 2. Leg B - The above call is routed through iCall and then answered by the >> same Freeswitch instance. >> 3. On Leg B, I play a file and attempt to get DTMF in a lua script. Here >> is the code of interest: >> >> callPin = session:playAndGetDigits(1, 10, 4, 30000, "#","/tmp/cw_17.wav", >> "", "\\d+"); >> >> 4. On Leg A, I send DTMF information in a lua script. Here is the code of >> interest. I initiate a delay between each digit: >> >> local newPin = ""; >> >> for i = 1, string.len(pin) do >> >> newPin = newPin .. string.sub(pin, i, i) .. "W"; >> >> end >> >> session:execute("send_dtmf", newPin .. "#@200"); >> >> ** Note that there is a session:sleep on Leg A before I send the DTMF to >> make sure i don't send it too early ** >> >> The problem is that the recognized DTMF on Leg B is wrong about 30% of the >> time. For example, if Leg A enters: 22063083, Leg B will get the DTMF digits >> 222063083. This is an example of duplication, but I have also experienced >> missing DTMF codes (and an occasional wrong code completely) >> >> I have messed with a bunch of DTMF settings in hopes of fixing this issue, >> but I cannot seem to find something that is reliable 100% of the time. >> >> _____________________________ >> >> Here are the DTMF settings I have looked at / messed with. I've tried >> various values for the dtmf-duration in the config (and in the send_dtmf >> command above) >> >>  >> >> <param name="dtmf-duration" value="500"/> >> >> <param name="dtmf-type" value="rfc2833"/> >> >> <param name="inbound-codec-prefs" value="$${global_codec_prefs}"/> >> >> <param name="outbound-codec-prefs" value="$${outbound_codec_prefs}"/> >> >> <param name="inbound-codec-negotiation" value="greedy"/> >>  >> <X-PRE-PROCESS cmd="set" data="global_codec_prefs=PCMU"/> >> <X-PRE-PROCESS cmd="set" data="outbound_codec_prefs=PCMU"/> >> ________________________ >> I have run fs_cli with event logging and the DTMF events that Freeswitch >> gets do correlate to the wrong value (e.g. the duplication / missing digits >> is noticable in the Freeswitch events as well). >> Also, I am not running any dtmf-related applications on the session before >> I give control to the lua scripts (e.g. not running start_dtmf) >> Has anyone experienced this type of issue? Or know what I can do to >> resolve it? My next step was going to be trying this against another >> provider besides iCall, but I figured I would see if anyone has encountered >> a similar problem before. Any help is appreciated. >> Thanks > > _______________________________________________ > FreeSWITCH-users mailing list > FreeSWITCH-users at lists.freeswitch.org > http://lists.freeswitch.org/mailman/listinfo/freeswitch-users > UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users > http://www.freeswitch.org > > -- Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire AIM: anthm MSN:anthony_minessale at hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale at gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888 at conference.freeswitch.org googletalk:conf+888 at conference.freeswitch.org pstn:+19193869900

2 1

Chat Event in Dingaling
by Sameer Khan 31 May '12

31 May '12

Is this implemented? There is no event fired when a chat message is received

2 2

uuid_record and recording output format
by Hector Geraldino 31 May '12

31 May '12

Greetings, I'm using a 3rd party application (ndev dragonmobile) to get the transcription of some audio recorded by FreeSWITCH. Think about it as a voicemail transcription service. The problem I'm facing is that, when I record a session using uuid_record, the output file is encoded in PCM 16-bit @ 8khz. Correct me if I'm wrong, but my understanding is that if I want to capture audio from calls coming from the PSTN (analog/landlines), the best I can do is to record it in 8-bits (using G.711). I don't want to use sox (or any other tool) to resample the output file, and what I've tried so far is setting the sample_rate variable on the diaplan as recommended on the wiki: http://wiki.freeswitch.org/wiki/Variable_record_rate <action application="set" data="record_sample_rate=8000"/> This doesn't have any effect on the generated wav file, which is still encoded in 16-bits. So my question is: does this variable affects the behavior of the uuid_record command? Or, do I really need to encode the audio output in 8-bits when the origin of the call comes from the PSTN? How is FreeSWITCH encoding the audio in 16-bits if, in theory, the best rate we can get from an analog line is 8-bits? Sorry if I'm misunderstanding something, but I'm not a telephony/voip guy, more like a java developer. Thanks for your help.

3 3

Access to Cause filed in Reason header
by Javier Gallart 31 May '12

31 May '12

Hello list some of our providers include a Reason header in negative replies to an INVITE. For instance, they might include a cause=65 inside a Sip 503 reply. We need to react upon some of those ISDN causes: in this particular case with if Reason header were absent we would simply do nothing but let FS to send the response upstream; but if cause=65 in present we want to resend the call to the same provider with a different set of offered codecs. Is there any way to access to the content of that header?. Our applicatiion listens to the event socket in inbound mode but I haven't found a way to do this. Any help would be appreciated. Regards Javi

2 1

SCA not working inbound - Multi Domain
by Sean Devoy 31 May '12

31 May '12

Hi all, I have a muti-tennnant configuration that is working nicely except for Shared Call Appearance. The desktop devices are CISCO 504Gs and they are configured as described in the FS Wiki as well as Cisco Documentation. The SCA works perfectly for outbound calls - if either phone pickups like 220, the other phones indicator light flashes red. However, inbound calls will go to only one of the phones (which one has changed a few times) and the other phones line still just stays green and does not ring. Here is the sip interfaces config: <profile name="internal"> <settings> <param name="enable-timer" value="false"/> <param name="user-agent-string" value="Configured by Sean!"/> <param name="rtp-timer-name" value="soft"/> <param name="codec-prefs" value="$${global_codec_prefs}"/> <param name="manage-shared-appearance" value="true"/> <param name="multiple-registrations" value="true"/> <param name="manage-presence" value="true"/> <param name="inbound-codec-negotiation" value="generous"/> <param name="inbound-reg-force-matching-username" value="true"/> <param name="nonce-ttl" value="86400"/> <param name="rfc2833-pt" value="101"/> <param name="manage-presence" value="true"/> <param name="auth-calls" value="true"/> <param name="sip-ip" value="10.10.40.185"/> <param name="rtp-ip" value="10.10.40.185"/> <param name="sip-port" value="5064"/> <param name="nat-options-ping" value="false"/> <param name="all-reg-options-ping" value="true"/> <param name="context" value="from-BFIS"/> </settings> </profile> The directory entry which both phones connect using: <user id="220"> <variables> <variable name="outbound_caller_id_name" value="BFIS Sean"/> <variable name="outbound_caller_id_number" value="410420BLEEP"/> <variable name="internal_caller_id_name" value="Sean BLEEP"/> <variable name="internal_caller_id_number" value="220"/> <variable name="user_context" value="from-internal-BFIS"/> <variable name="user_originated" value="true"/> <variable name="toll_allow" value="domestic"/> <variable name="accountcode" value="220"/> <variable name="mwi-account" value="220@voicemail_BFIS"/> </variables> <params> <param name="manage-shared-appearance" value-="true" />" <param name="password" value="BLEEPBLEEP"/> <param name="dial-string" value="{presence_id=220@mydomainname.com}${sofia_contact(220@mydomainname.co m)}"/> <param name="mwi-account" value="220@voicemail_BFIS"/> </params> </user> And the dial plan for ext 220: <extension name="220" > <condition field="destination_number" expression="^220$"> <action application="set" data="effective_caller_id_number=${internal_caller_id_number}"/> <action application="set" data="effective_caller_id_name=${internal_caller_id_name}"/> <action application="set" data="ringback=${us-ring}"/> <action application="set" data="call_timeout=20"/> <action application="set" data="hangup_after_bridge=true"/> <action application="bridge" data="{sip_invite_domain=mydomainname.com}user/220(a)mydomainname.com" /> <action application="answer"/> <action application="voicemail" data="default voicemail_BFIS 220"/> <action application="hangup"/> </condition> </extension> I did see this in the wiki (http://wiki.freeswitch.org/wiki/Shared_Line_Appearance) If SLA works for outgoing calls and SLA does not work for inbound calls to the SLA phones, you may have some presence problem related to mixed IP and domain names. When using ODBC you may issue the following SQL statement select sip_to_host,sip_from_user,sip_from_host,hostname,presence_id,call_info from sip_dialogs; But I don't have ODBC on this server, so I am a little lost. I have the phones login to domain names, not addresses. I never refer to IP addresses in my xml (except gateways addresses). I am not trying SLA across domain, only within the same domain. I hope someone can spot something. Thanks for your help. Sean

3 10