- FreeSWITCH-users - FreeSWITCH Mailing Lists

mod_dingaling segfault with libgcrypt and gnutls on Fedora 13
by Jeffrey Leung 08 Oct '10

08 Oct '10

Hey all, I've read and understand somewhere that with a bad copy of gnutls and libgcrypt can cause mod_dingaling to segfault when loading or running. However, I'd like to ask if there's anyone out there trying to use mod_dingaling with Fedora 13's included libgcrypt and gnutls-devel packages as I am experiencing segfaults when I try to load mod_dingaling.

1 0

FreeSWITCH parameters for voice quality
by rex.alex 08 Oct '10

08 Oct '10

Hello, I have FreeSWITCH on Cent OS 5.4 physical server and I am using X-Lite soft phone on Windows client to make external calling through a registered gateway. But, I am facing issues like voice breakage and static noise on calls. I am monitoring the network connectivity but before that I want to make sure that all FreeSWITCH parameters have been configured correctly. Please let me know as far as FreeSWITCH is concern what are the parameters that I have to check for these kind of issues? Thanks, Rex -- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/FreeSWITCH-parameters-for-voi… Sent from the freeswitch-users mailing list archive at Nabble.com.

3 2

Freeswitch - mod_spidermonkey_curl bug?
by Shamun toha md 08 Oct '10

08 Oct '10

Hello, I have a very strange problem using mod_spidermonkey_curl in CentOS 5.5 x86_64 bit. Same setup i am using in other machine in Fedora 12, it works, is there any bugs????????? I receive OK, but when i use to compare it fails in logical statement. FreeSWITCH Version 1.0.head (git-25ddef5 2010-09-24 01-55-54 +0200) Started. Installation curl as: ============== [root@a autoload_configs]# cat spidermonkey.conf.xml <configuration name="spidermonkey.conf" description="Spider Monkey JavaScript Plug-Ins"> <modules> <load module="mod_spidermonkey_teletone"/> <load module="mod_spidermonkey_core_db"/> <load module="mod_spidermonkey_socket"/>  * <load module="mod_spidermonkey_curl" />* </modules> </configuration> .../scripts/helloworld.js: ============ use("CURL"); var menuselection = ""; var promptfordigits_dtmf_digits = ""; var invalidRetry = 0; /* This is getting failed, in Centos, but in other machine it works */ function my_callback(string, arg) { console_log("info", string); //string = 'OK'; // then it works!! //2010-10-08 18:40:53.585676 [INFO] mod_dialplan_xml.c:331 Processing 1002 <1002>->1200 in context default * //2010-10-08 18:41:23.842965 [INFO] helloworld.js:42 OK* //2010-10-08 18:41:35.962099 [NOTICE] switch_ivr.c:1480 Transfer sofia/internal/1002(a)z.com to XML[3000@default] //2010-10-08 18:41:35.962099 [INFO] mod_dialplan_xml.c:331 Processing 1002 <1002>->3000 in context default //2010-10-08 18:41:43.134207 [NOTICE] sofia.c:528 Hangup sofia/internal/ z.com [CS_EXECUTE] [NORMAL_CLEARING] if (string=='OK') { // // // <<<<<<<<<<<<<<<<<<------------------ It receive OK but it never execute this two, and but it execute invalid macro // // // session.sayPhrase("valid", menuselection, "en"); session.execute("transfer", '3000'); } else { session.sayPhrase("invalid", menuselection, "en"); return true; } /* FS wiki - example */ function promptfordigits_dtmf_callback(session, type, digits, arg) { console_log("digit: " + digits.digit + "\n"); promptfordigits_dtmf_digits += digits.digit; /* returning true does not interrupt the digit collection*/ return(true); } /* FS wiki - example */ function promptfordigits(ivrsession, promptname, numdigits, timeout) { var repeat = 0; console_log("saymenu: menu=[" + promptname + "] numdigits=[" + numdigits + "]\n"); session.flushDigits(); promptfordigits_dtmf_digits = ""; while (ivrsession.ready() && promptfordigits_dtmf_digits.length < numdigits && repeat < 3) { /* play phrase - if digit keyed while playing callback will catch them. If less than numdigits collected we get the rest after the prompt.*/ ivrsession.sayPhrase(promptname, numdigits, "en", promptfordigits_dtmf_callback, ""); console_log("Prompt done=[" + promptname + "] Collected " + promptfordigits_dtmf_digits.length + " digits [" + promptfordigits_dtmf_digits + "]\n"); /* if caller still here and has not entered any selection yet (or less than numdigits entered) - wait for the rest of the digits*/ if (ivrsession.ready() && promptfordigits_dtmf_digits.length < numdigits ) { promptfordigits_dtmf_digits += ivrsession.getDigits(numdigits - promptfordigits_dtmf_digits.length, "", timeout); /* if still no selection or insufficient digits repeat menu */ if (promptfordigits_dtmf_digits.length < numdigits) { promptfordigits_dtmf_digits = ""; repeat++; } } } return(promptfordigits_dtmf_digits); } function retry() { /* FS wiki - example */ menuselection = promptfordigits(session, "DialByNumberMenu", 8, 6000); if ( session.ready() ) { var curl = new CURL(); curl.run( "POST", "http://portal/conference/query", "om=" + menuselection + "&ty=__", my_callback, "my arg\n", "p:ier"); } } /** * Let's answer our call */ session.answer(); retry(); Thanks & Regards

1 0

FS starts re-registering every minute after a few hours
by Mario 08 Oct '10

08 Oct '10

I really could use some help on this one, this is one of the last major issues before I switch to FS. I have expire-seconds set at 600, this works for several hours. Then, for no apparent reason it changes to 60 (see trace). Most calls are missed due to this issue. Why would FS change it and how can I keep it at 600? BTW retry is at 30. I spent lot's of time with traces/investigating and the ITSP watched it and says they are not changing anything. Thanks! Mario PS If it's them how do I prove it? send 865 bytes to udp/[111.222.1.36]:5060 at 15:07:22.220308: ------------------------------------------------------------------------ REGISTER sip:sip.itsp1.com SIP/2.0 Via: SIP/2.0/UDP 100.101.102.103:5080;rport;branch=hideBcm183mmt2UD Max-Forwards: 70 From: <sip:uuid1@sip.itsp1.com;transport=udp>;tag=hide37BeUD To: <sip:uuid1@sip.itsp1.com;transport=udp> Call-ID: hide-a069-49d0-b342-hide34d28 CSeq: 2914438 REGISTER Contact: <sip:gw+uuid1@100.101.102.103:5080;transport=udp;gw=uuid1> Expires: 60 User-Agent: FreeSWITCH-mod_sofia/1.0.head-git-f499c0c 2010-10-07 10-27-31 -0500 Allow: INVITE, ACK, BYE, CANCEL, OPTIONS, MESSAGE, UPDATE, INFO, REGISTER, REFER, NOTIFY Supported: timer, precondition, path, replaces Proxy-Authorization: Digest username="uuid1", realm="xxxtel.com", nonce="hide0000123270bed298a5fee464c4bb367750hide", algorithm=MD5, uri="sip:sip.itsp1.com", response="70000a710c1b7a363899922e6hide" Content-Length: 0 ------------------------------------------------------------------------ recv 567 bytes from udp/[111.222.1.36]:5060 at 15:07:22.321019: ------------------------------------------------------------------------ SIP/2.0 407 Proxy Authentication Required Via: SIP/2.0/UDP 100.101.102.103:5080;rport=5080;branch=hideBcm183mmt2UD;received=111.222.333.127 From: <sip:uuid1@sip.itsp1.com;transport=udp>;tag=hide37BeUD To: <sip:uuid1@sip.itsp1.com;transport=udp>;tag=9214944d68c027e4c99a14a397962871.4d99 Call-ID: hide-a069-49d0-b342-hide34d28 CSeq: 2914438 REGISTER Proxy-Authenticate: Digest realm="xxxtel.com", nonce="4caf33ab99012dadc6e7c84040e1b4ff8841a343edc5b0c3", stale=true Server: OpenSIPS (1.6.3-notls (i386/freebsd)) Content-Length: 0 ------------------------------------------------------------------------ send 865 bytes to udp/[111.222.1.36]:5060 at 15:07:22.321391: ------------------------------------------------------------------------ REGISTER sip:sip.itsp1.com SIP/2.0 Via: SIP/2.0/UDP 100.101.102.103:5080;rport;branch=z9hG4b99m5c33mrH3reS Max-Forwards: 70 From: <sip:uuid1@sip.itsp1.com;transport=udp>;tag=hide37BeUD To: <sip:uuid1@sip.itsp1.com;transport=udp> Call-ID: hide-a069-49d0-b342-hide34d28 CSeq: 2914439 REGISTER Contact: <sip:gw+uuid1@100.101.102.103:5080;transport=udp;gw=uuid1> Expires: 60 User-Agent: FreeSWITCH-mod_sofia/1.0.head-git-f499c0c 2010-10-07 10-27-31 -0500 Allow: INVITE, ACK, BYE, CANCEL, OPTIONS, MESSAGE, UPDATE, INFO, REGISTER, REFER, NOTIFY Supported: timer, precondition, path, replaces Proxy-Authorization: Digest username="uuid1", realm="xxxtel.com", nonce="4caf33ab00012dadc6e7c89940e1b4ff9941a343edc5b0c3", algorithm=MD5, uri="sip:sip.itsp1.com", response="b8597203297d6db518b9996f55d3dc36" Content-Length: 0 ------------------------------------------------------------------------ recv 503 bytes from udp/[111.222.1.36]:5060 at 15:07:22.421206: ------------------------------------------------------------------------ SIP/2.0 200 OK Via: SIP/2.0/UDP 100.101.102.103:5080;rport=5080;branch=z9hG4bKKm5c999rH3reS;received=111.222.333.127 From: <sip:uuid1@sip.itsp1.com;transport=udp>;tag=hide37BeUD To: <sip:uuid1@sip.itsp1.com;transport=udp>;tag=921495599c027e4c99a14a397962871.9262 Call-ID: hide-a069-49d0-b342-hide34d28 CSeq: 2914439 REGISTER Contact: <sip:gw+uuid1@111.222.333.127:5080;transport=udp;gw=uuid1>;expires=60 Server: OpenSIPS (1.6.3-notls (i386/freebsd)) Content-Length: 0

1 0

bypass_media only if both peers have public IP: how to do it?
by Fabio Pietrosanti (naif) 08 Oct '10

08 Oct '10

Hi, i would like to configure a FS system so that only if both peers have a public IP address they can speak each other. The condition is that the client are mobile and can connect from whatever network, so apply-nat-acl does not work because some mobile carrier use fancy internal IP address such as 1.1.1.1 . All clients connect with SIP/TLS and there is no SIP ALG that mangle the SDP, so we can be sure that what FS see is exactly what the SIP client has sent. Given that, i would like FS to always do proxy_media unless BOTH two leg of the call have a public IP address. I know that's not extremely efficient and that in theory two SIP clients with one behind NAT and one with public IP can speak each other, but even in such conditions there may be unpredictable issues. How the NAT handling of FS could be modified to do that and let bypass_media only when both SIP client have a public IP address? Fabio

1 0

segfault in swift library
by covici＠ccs.covici.com 08 Oct '10

08 Oct '10

I have been trying to get mod_cepstral to work and so far, no joy at all. At first I was using the version 5 voice, but was told on the IRC or maybe in the onference that I should go back to version 4. So I restore the directory, recompiled FSW and now I see in the logs the following message: Oct 7 06:16:15 ccs kernel: freeswitch[10398]: segfault at 627769 ip b5ffda43 sp bfd48ce4 error 4 in libswift.so.4.2[b5fcf000+7e000] No core was dumped -- can someone tell me what is going on, or how to troubleshoot this? Thanks in advance for any suggestions. -- Your life is like a penny. You're going to lose it. The question is: How do you spend it? John Covici covici(a)ccs.covici.com

1 0

Possible bug with bypass_media=true functionality
by Pete Kelly 08 Oct '10

08 Oct '10

Hi, I posted a question to this list yesterday because I was having problems with my calls ending as soon as they were bridged. I was originating 2 outbound calls from a lua script, then bridging them. I managed to get round the problem of the calls shutting down immediately by initiating calls using the originate command, thusly: originate sofia/lpmedia/2000(a)10.15.20.122 '&lua(click2call_call.lua)' This causes the lua script to initiate with a session already present and then a bridge to a new call (legB) completes successfully. Within the lua script I am bridging to leg B like this: session:execute("bridge", "[bypass_media_after_bridge=true]sofia/lpmedia/ 91979197(a)10.15.20.122") This works, but the reINVITEs which get sent out after legB is established contain the wrong sip request URI. - The SIP URI in the reINVITE to Leg A is to the MEDIA IP of leg B (the SIP packet is sent to the correct IP, just the request URI contains the media IP) - The SIP URI in the reINVITE to Leg B is to the MEDIA IP of leg A... again the SIP packet is sent to the correct IP, it's just the request URI which contains the media IP. In some instances the request URI doesn't contain the username either, it is just INVITE sip:<MEDIA_IP> I am using the latest sources of freeswitch from git Is this a bug I need to report? It is causing me to receive 404 and 402 errors on the reinvite as the request URI is invalid. Pete

3 4

Fax hint request
by Ale 08 Oct '10

08 Oct '10

Hi all, I've compiled freeswitch from git, configured an external trunk and added enabled-t38=true in fax.conf. Both servers have a public ip. Tring to send a fax from cli via origiante all t.38 fax fail with 2010-10-05 19:48:14.228734 [NOTICE] mod_sofia.c:880 Hangup sofia/external/xxxxxx@ip [CS_EXECUTE] [RECOVERY_ON_TIMER_EXPIRE] ... 2010-10-05 19:48:14.228734 [DEBUG] mod_spandsp_fax.c:335 Fax processing not successful - result (49) The call dropped prematurely. Reading documentation i've found that RECOVERY_ON_TIMER_EXPIRE can be caused by nat configuration problem, but i think configuration could be reasonably correct. sip flow: http://i56.tinypic.com/30j1ezn.png remote trying after the initial invite of fs: http://i52.tinypic.com/255hlps.png remote 183 session in progress: http://i54.tinypic.com/293jew2.png After 183 fs wait some time, probably a timeout, and send a cancel request. On fs cli i see also the data below, but fs never send to remote trunk. 2010-10-05 19:48:04.235005 [DEBUG] sofia_glue.c:137 sofia/external/xxxx@ip image media sdp: v=0 o=FreeSWITCH 1286300921 1286300923 IN IP4 213.187.9.30 s=FreeSWITCH c=IN IP4 xxx.xxx.9.30 t=0 0 m=image 21562 udptl t38 a=T38FaxVersion:0 a=T38MaxBitRate:14400 a=T38FaxFillBitRemoval a=T38FaxRateManagement:transferredTCF a=T38FaxMaxBuffer:2000 a=T38FaxMaxDatagram:400 a=T38FaxUdpEC:t38UDPRedundancy Can anyone give me a hint, if i'm wrong something? Many thx, Ale

2 4

Can shared mailbox signal multiple MWIs?
by Mario 08 Oct '10

08 Oct '10

If possible, how can FS light multiple userid/phones NOT using multiple registrations MWI when a message is in a shared mailbox? I have had a lot of trouble with multiple registrations and can't get SLA/SCA to work on the SPA962s. Use of :_: was required in the bridge and that was a lot of headaches. All due to use of shared mailboxes. I can control what mailbox is played back via an extension but I can't find a way to have the MWI indicator come on multiple userid/phones for a shared mailbox on SPA962s. If I can get past this I can finally dump shared lines, they were also a pain on the SPA9000 I am trying to replace. Thanks! Mario

1 0

Question about effective caller ID & origination caller ID
by Mario 08 Oct '10

08 Oct '10

I have the inbound caller ID working for a bridge but can't reconcile why effective_caller_id.. does not work. I spent a lot of time the last few days reading the wiki and other web info. Since I can't find the answer anywhere can anyone shed some light on this? The wiki states: effective_caller_id_number Sets the effective callerid number. This is automatically exported to the B-leg; however, it is not valid in an origination string. In other words, set this before calling bridge, otherwise use origination_caller_id_number Does NOT work before the bridge:  Does NOT work before the bridge:  <!--action application="set" ONLY way IT WORKS (must be added before dialstring): <action application="bridge" data="<originate_timeout=15,origination_caller_id_number=${caller_id_number},origination_caller_id_name=${caller_id_name}>group/MGT@${domain_name}:_:sofia/gateway/uuid1/19162223333:_:sofia/gateway/uuid1/19164445555"/> The caller ID is lost in an extension, although I still don't know why at least I got it back.

3 4