- FreeSWITCH-users - FreeSWITCH Mailing Lists

Adding new Gateway with registration false - incoming calls works but outgoing calls not
by K K 01 Jul '14

01 Jul '14

Hello all, I have the following problem. I have added a new Gateway (SIP Trunk) on my FS simply by adding new configuration file in the "external" folder, new dialplan file in the "dialplan\default" folder and adding the CIDR of my Gateway as the allowed domain in the acl.conf.xml. And it works perfectly. Then I have change my configuration - I have added the "register false" parameter into my Gateway configuration and the result is that I am not longer able to call any of my Gateway SIP endpoint from FS, but FS is still able to receive the incoming calls from the Gateway, so I have lost the outgoing calls. I wondering if such configuration is in fact supported? I have 2 book about FS and in both of them there are only examples with registration set to true. I was looking in google for some help, but I have only found that I need to set the call authentication to false - but this was my default setting actually. Thank you for help in that topic.

1 0

Problems with RTP delay at around 300 concurrent calls
by D D 30 Jun '14

30 Jun '14

Hi, I would like to put 500 - 1000 calls into a FS system. I have a Dell R900 server with these properties: 8 processors at 2.40GHz 8GB RAM Gigabit ethernet I use sipp to send RTP streams. I use the FS settings recommended here: https://wiki.freeswitch.org/wiki/Performance_testing_and_configurations including the in-memory core.db. I also have used the "-rp" and "-nosql" settings on the FS command-line, and also used the "chrt" command to increase the priority to real-time. In all cases, at around 300 concurrent calls, the RTP inter-packet delay becomes unacceptably high. The voice quality is severely impacted. During the test, the CPU is not overloaded. It is using only about 1.5 of the 8 processors. Are these results expected? Has anyone gone higher than 300 concurrent calls? Is there anything else that I can do with configuration to improve throughput? Thanks, David

6 15

send_dtmf not working
by Matt Broad 30 Jun '14

30 Jun '14

Hi, I have the need to send a dtmf tone, but send_dtmf does not seem to work. I have the following in a dialplan, but doing a sip trace the dtmf digits are not sent. I have tried removing the start_dtmf and start_dtmf_generate, but still no joy. <action application="start_dtmf" /> <action application="start_dtmf_generate"/> <action application="answer"/> <action application="sleep" data="2000"/> <action application="send_dtmf" data="0123456789ABCD*#@100"/> <action application="sleep" data="2000"/> I am using FreeSWITCH Version 1.4.6+git~20140603T193516Z~9479729340~64bit (git 9479729 2014-06-03 19:35:16Z 64bit) thanks Matt

2 2

P-Asserted-Identity header issue
by Ričardas Stoma 30 Jun '14

30 Jun '14

Hello guys, i have a problem when i try to read P-Asserted-Identity header. Let's say i send to freeswitch this header: P-Asserted-Identity: <sip:123456789@192.168.0.73:5060;user=phone> But when i try to read this header in my dialplan, i get: 123456789 When i check channel variables i see: variable_sip_P-Asserted-Identity: 370370370370 maybe there is a way to read full P-Asserted-Identity header? I need to resend this header without modifying it. Is it possible?

1 0

LUA BUG not updating the hangup cause once originated the call
by Ashou Han 30 Jun '14

30 Jun '14

Hi there, There is a bug for Lua, that the hangup cause not updated once originated the call. I originated a call to my opensips proxy with a 183 back, lua hangupCause return SUCCESS, but few seconds later the far end carrier return USER_BUSY (see this in the DEBUG log). But session:hangupCause() still return SUCCESS, not the updated cause: USER_BUSY. FS version: FreeSWITCH Version 1.2.22+git~20140309T154101Z~455941d369~64bit (git 455941d 2014-03-09 15:41:01Z 64bit) FreeSWITCH Version 1.2.23+git~20140506T190611Z~88216387e0~64bit (git 8821638 2014-05-06 19:06:11Z 64bit) FreeSWITCH Version 1.2.10+git~20130628T211259Z~e29731d673 (git e29731d 2013-06-28 21:12:59Z) All have the same bug! Thanks, Ashou Han

2 2

Multiple inbound ACL on external profile
by Joel White 30 Jun '14

30 Jun '14

Is it possible to multiple inbound acl lists on a single profile?

2 2

set variable on first leg when bridge is answered
by Grant Bagdasarian 30 Jun '14

30 Jun '14

Hello, I'm first doing a originate which hits the below dialplan when answered. When it's answered I'm setting a variable named "bridged" to 0. When the user presses the "critical_option_digit" the call is bridged to another destination. I need to know if this bridge was answered or not. I tried different ways of execute_on_answer but that didn't work: <action application="set" data="execute_on_answer=set bridged=1"/> <action application="set" data="nolocal:execute_on_answer=set bridged=1"/> <action application="export" data="execute_on_answer=set bridged=1"/> <action application="export" data="nolocal:execute_on_answer=set bridged=1"/> It is important the bridged variable is set to 1 on the first leg and a header is appended to the BYE/200 OK(BYE) of the first leg containing this new value. Any ideas how to fix this? <?xml version="1.0" encoding="utf-8"?> <include> <context name="outbound_cc"> <extension name="start"> <condition field="destination_number" expression="^(start)$" require-nested="false"> <action application="answer" /> <action application="set" data="bridge_early_media=true" /> <action application="set" data="hangup_after_bridge=false" /> <action application="set" data="critical_option_digit=1" /> <action application="set" data="transfer_destination=0031123456789" /> <action application="set" data="bridged=0" />  <action application="play_and_get_digits" data="1 1 2 5000 # welcome.wav invalid.wav dtmfInput \d 2500"/> <action application="transfer" data="act_on_input XML outbound_cc" /> </condition> </extension> <extension name="act_on_input"> <condition field="${dtmfInput}" expression="${critical_option_digit}" break="never">  <action application="set" data="execute_on_answer=set bridged=1"/> <action application="bridge" data="{sip_copy_custom_headers=false}sofia/outbound/${transfer_destination}@$${outbound_load_balancer_sip_ip}:$${outbound_load_balancer_sip_port}" /> <action application="export" data="sip_bye_h_X-CollectedData=<CollectedData><DTMFInput>${dtmfInput}</DTMFInput><BridgeSucceeded>${bridged}</BridgeSucceeded></CollectedData>"/> <action application="export" data="sip_rh_X-CollectedData=<CollectedData><DTMFInput>${dtmfInput}</DTMFInput><BridgeSucceeded>${bridged}</BridgeSucceeded></CollectedData>"/>  <anit-action application="export" data="sip_bye_h_X-CollectedData=<CollectedData><DTMFInput>${dtmfInput}</DTMFInput><BridgeSucceeded>${bridged}</BridgeSucceeded></CollectedData>"/> <anit-action application="export" data="sip_rh_X-CollectedData=<CollectedData><DTMFInput>${dtmfInput}</DTMFInput><BridgeSucceeded>${bridged}</BridgeSucceeded></CollectedData>"/> <anti-action application="hangup"/> </condition> </extension> </context> </include>

2 2

failed to play audio
by Joel Dodson 30 Jun '14

30 Jun '14

Hi, I'm having intermittent trouble playing audio in a call. The scenario is: - there's an incoming call which triggers an outgoing call. - Audio is played to the incoming leg while the outgoing leg is ringing. - Once the outgoing call is connected, audio on the incoming legs is stopped and the legs are bridged - recording is started on the inbound call leg - audio is being played on the outgoing leg while the legs are being bridged. This is all done using event sockets, not any of the advanced features of the XML dialing plan. The problem is intermittent and hard to reproduce (of course, it's easily reproducible in our production environment by a customer...). I have found a significant difference in logs on calls where the problem has occurred. On the problem call, I don't see a call made to mod_httapi to cache the audio file to be played. Here are some logs for a call with the problem (newer logs are at the top, I extracted these logs from a logging aggregation tool) : 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:483 (c-4fkbyrefwy3jgl2e5kfzori) State SOFT_EXECUTE" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:415 (c-4fkbyrefwy3jgl2e5kfzori) Running State Change CS_SOFT_EXECUTE" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:473 (c-4fkbyrefwy3jgl2e5kfzori) State RESET going to sleep" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_session.c:1333 Send signal c-4fkbyrefwy3jgl2e5kfzori [BREAK]" 2014-06-26 19:57:59.604394 [DEBUG] switch_ivr_bridge.c:764 (c-4fkbyrefwy3jgl2e5kfzori) State Change CS_RESET -> CS_SOFT_EXECUTE"2014-06-26 19:57:59.604394 [DEBUG] switch_ivr_bridge.c:757 c-4fkbyrefwy3jgl2e5kfzori CUSTOM RESET" 2014-06-26 19:57:59.604394 [DEBUG] mod_sofia.c:168 c-4fkbyrefwy3jgl2e5kfzori SOFIA RESET" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:473 (c-4fkbyrefwy3jgl2e5kfzori) State RESET" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:415 (c-4fkbyrefwy3jgl2e5kfzori) Running State Change CS_RESET" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:492 (c-4fkbyrefwy3jgl2e5kfzori) State HIBERNATE going to sleep" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_session.c:1333 Send signal c-4fkbyrefwy3jgl2e5kfzori [BREAK]" 2014-06-26 19:57:59.604394 [DEBUG] switch_ivr_bridge.c:772 (c-4fkbyrefwy3jgl2e5kfzori) State Change CS_HIBERNATE -> CS_RESET" 2014-06-26 19:57:59.604394 [DEBUG] mod_sofia.c:224 c-4fkbyrefwy3jgl2e5kfzori SOFIA HIBERNATE" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:492 (c-4fkbyrefwy3jgl2e5kfzori) State HIBERNATE" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:415 (c-4fkbyrefwy3jgl2e5kfzori) Running State Change CS_HIBERNATE" 2014-06-26 19:57:59.604394 [DEBUG] switch_core_state_machine.c:477 (c-4fkbyrefwy3jgl2e5kfzori) State EXECUTE going to sleep"2014-06-26 19:57:59.604394 [DEBUG] switch_ivr.c:612 c-4fkbyrefwy3jgl2e5kfzori Command Execute playback({tagId=tag-6gtqkkwg62k7xziefiyv7vq,}http://(nohead=true,ext=mp3)www.t tsprovider.com/tts/gen.php?EID=2&LID=1&VID=1&TXT=Call+from+Direct&EXT=mp3&A…) EXECUTE c-4fkbyrefwy3jgl2e5kfzori playback({ta gId=tag-6gtqkkwg62k7xziefiyv7vq,}http://(nohead=true,ext=mp3) www.ttsprovider.com/tts/gen.php?EID=2&LID=1&VID=1&TXT=Call+from+Direct&EXT=… CS=abcde0123456789)" 2014-06-26 19:57:59.534396 [DEBUG] sofia.c:7541 dispatched freeswitch event for INFO" 2014-06-26 19:57:59.534396 [DEBUG] switch_core_session.c:915 Send signal c-4fkbyrefwy3jgl2e5kfzori [BREAK]" and here's what we see in the general case (these logs are directly from freeswitch.log so newest logs at the top): 2014-06-28 21:22:13.086522 [DEBUG] switch_ivr.c:612 c-dvji55mpysr3gcyruol6fgi Command Execute playback(http:// (nohead=true,ext=mp3) www.ttsprovider.com/tts/gen.php?EID=2&LID=1&VID=1&TXT=Hi+guys&EXT=mp3&ACC=X… ) EXECUTE c-dvji55mpysr3gcyruol6fgi playback(http://(nohead=true,ext=mp3) www.ttsprovider.com/tts/gen.php?EID=2&LID=1&VID=1&TXT=Hi+guys&EXT=mp3&ACC=X… ) 2014-06-28 21:22:13.086522 [DEBUG] switch_event.c:1608 Parsing variable [nohead]=[true] 2014-06-28 21:22:13.086522 [DEBUG] switch_event.c:1608 Parsing variable [ext]=[mp3] 2014-06-28 21:22:13.446522 [DEBUG] mod_httapi.c:2482 caching: url: http://www.ttsprovider.com/tts/gen.php?EID=2&LID=1&VID=1&TXT=Hi+guys&EXT=mp… to /var/lib/freeswitch/storage/http_file_cache/e0aaab89166c3ebf4391ec147a1d0b2a.mp3 (8198 bytes) 2014-06-28 21:22:13.446522 [DEBUG] switch_ivr_play_say.c:1308 Codec Activated L16@8000hz 1 channels 20ms 2014-06-28 21:22:14.546521 [DEBUG] switch_ivr_play_say.c:1693 done playing file http://(nohead=true,ext=mp3) www.ttsprovider.com/tts/gen.php?EID=2&LID=1&VID=1&TXT=Hi+Avery+guys&EXT=mp3… As you can see, in the error case, instead of caching the the audio from our tts provider, the channel transitions immediately into HIBERNATE state. We're using the Java Event Socket Client ( https://wiki.freeswitch.org/wiki/Java_ESL_Client) to send/receive commands and events to/from freeswitch. We're using version 0.9.2 directly from the maven repository. And here's our freeswitch version: FreeSWITCH version: 1.2.8+git~20130430T234615Z~95d31525ce (git 95d3152 2013-04-30 23:46:15Z) I've been digging into this for quite some time. I'm concerned we're doing a lot of stuff in parallel that might be causing problems internally in freeswitch. I'd be very grateful for any suggestions on what's happening and how we can avoid it. thanks, Joel

2 1

Disable ipv6 altogether?
by Doug Vought 30 Jun '14

30 Jun '14

Is there a way to disable a gateway from ipv6 altogether? I'm having problems with registration that I think may be caused by multiple ipv4/ipv6 registrations to the same host.

2 1

gsmopen remote hangup/no answer detection issue
by Dmitry Sytchev 30 Jun '14

30 Jun '14

When I originate call with gsmopen and Huawei e169 dongle, call gets delivered to callee, and if he drops call, modem doesn't get hangup notification or busy indication. There are no difference whether call was answered or not, when calle hangs up, modem on our side continues to ring or just to send media from other side. Maybe there should be some special AT commands or something operator-specific to make this work? There is no messages from modem in logs except these after originate timeout hangup. I haven't found any issues like this in google and this list, so maybe its just our misconfiguration... Should I file a JIRA? -- Best regards, Dmitry Sytchev, IT Engineer

1 1