- FreeSWITCH-users - FreeSWITCH Mailing Lists

SSL implementation in FreeSWITCH
by Emrah 16 Mar '16

16 Mar '16

Hi all, I’m writing to document where I’m at with my issues with FreeSWITCH and SSL / TLS and share my conclusions so far. I am hoping that this can give lieu to some further testing in different environments, and a proper fix if a bug is indeed confirmed. First, I am running FreeSWITCH 1.6.6 on a Debian 8. Vars.xml shows sip_tls_version=tlsv1,tlsv1.1,tlsv1.2. What I’ve observed is that in a sequence where client sens an invite to FS; FS responds with 407 proxy authorization required; client sends ack; Client sends the invite with the digest auth. The last packet can easily exceed the max segment size of a TCP segment, typically if the SDP advertises a bunch of codecs, or if the client uses SRTP and the SAVP contains many crypto suites. Now, when this occurs, the packets should be sent fragmented so they can fit in the MTU. It is then up to the receiving end to reassemble the segments and feed the complete packet to the application layer. What I’ve noticed is that a packet that is too large is simply never received by FreeSWITCH. Since this is systematically the case with every software and hardware client I’ve used, I am drawn to think that the issue lies in the SSL implementation of FreeSWITCH. In the event that for some reason my network or server OS configuration may be behind this, I would appreciate if someone would be willing to share some SIP credentials that can let me test TLS and SRTP. If getting to the bottom of this is of interest to any of you, I’d obviously be keen on handing out a couple of accounts. I hope this message can be the starting point of a fruitful resolution process. Thank you if you’ve read this up to here. Now hit reply and give me your 2 cents! :) Best, Emrah

6 10

Freeswitch IVR: call drop on transfer to busy extensions
by Francesco Piccinin 15 Mar '16

15 Mar '16

Hello Giovanni, thanks for you quick answer. Please find below patebin link you requested: XML Dialplan file's extract: http://pastebin.com/dmHAKWxT Internal call fscli logs: http://pastebin.com/KfnaYN11 External call fscli logs http://pastebin.com/2Khp1MaX Regarding dialplan files, I'm going to explain here call flow: Default.xml 1.call hits extension call_local, lua script is run and channel vasr cType=SRV, service_type=IVR 2. extension apps in xml services is execute Services.xml lua script is run to retrieve ivr name 3. on selection (2 in the logs) call is transfer to 8910 XML calls Calls.xml 4. extension: call_inside is execute and call is hangup after <action application="limit" data="db ${domain_name} MAX_${cCalled} ${maxcallsB} !USER_BUSY"/> As you can see from the logs mod sofia hangup with user busy and then send a bye to external or internal channel: 2016-03-14 11:37:40.266607 [DEBUG] mod_sofia.c:506 Channel sofia/external/ 00403737752(a)10.200.24.10 hanging up, cause: USER_BUSY 2016-03-14 11:37:40.266607 [DEBUG] mod_sofia.c:558 Sending BYE to sofia/external/00403737752(a)10.200.24.10 2016-03-14 11:38:54.166606 [DEBUG] mod_sofia.c:506 Channel sofia/internal/ 5532(a)insiel.fvgvoipcoll.it hanging up, cause: USER_BUSY 2016-03-14 11:38:54.166606 [DEBUG] mod_sofia.c:558 Sending BYE to sofia/internal/5532(a)insiel.fvgvoipcoll.it I tried to removed the limit app and the calls then hint limit_exceeded extension in calls.xml but behaviour is the same. It seems to me that continue_on_fail where passing over IVR is not considered. If I call directly (without IVR) extension 8910 (while it is busy) behaviour for both internal or external call is correct and busy tone is received. Looking forward for a clue :-) Thanks Regards Francesco

3 2

SILK codec
by Oleg Stolyar 15 Mar '16

15 Mar '16

Hi guys, does FreeSWITCH still support SILK?

5 17

freeswitch with tickless linux kernel
by François 15 Mar '16

15 Mar '16

Is it still recommended to use custom 1000HZ kernels for FS servers or can we go tickless now? Any difference if FS is installed in baremetal vs virtualized? Thanks, François.

5 6

socket_resume + bridge + att_xfer
by Luiz Fernando Softov 15 Mar '16

15 Mar '16

Hi, i'm using dialplan socket (outbound), until now it's ok. But sometimes if my daemon crash (i'm working on it) the call still active. Searching i found socket_resume "If this variable is set to true, the dialplan will resume execution with the next action after the call to the socket application. This can be used for example to allow you to do something intelligent in the dialplan if your IVR application gets killed in an unclean way. If there is a bridge active when the disconnect happens, it is killed. To do this from your application after the socket is already connected, issue the resume command." So i make <action application="set" data="socket_resume=true"/> <action application="socket" data="127.0.0.1:3012 async full"/> <action application="hangup"/> Now, when i make a bridge (A -> B), and make a att_xfer (B -> C), when B press 0 (tree_way), A hangup. If i remove socket_resume, it's OK (A+B+C), but if i kill my daemon (simulate a crash), the call still ACTIVE. Is there another way to do that? There is something I'm missing?

1 0

FS, SS7 and Dialogic IMG2020 in the UK
by Gavin Henry 15 Mar '16

15 Mar '16

Hi all, We're looking to get a pair of Dialogic IMG2020 for some BT Ofcom regulated SS7 interconnects: http://www.dialogic.com/en/products/gateways/img/img2020.aspx Any issues known with FS behind these? Thanks, Gavin. -- http://www.surevoip.co.uk

6 7

opensips+freeswitch on different amazon ec2 servers: 488 error
by Стас Тельнов 14 Mar '16

14 Mar '16

Hello. I have had difficulties at control of the opensips and freeswitch connective. The task is: to organize voip-conference between mobile application users through our opensips server or between mobile application users with addition of a call to the city line. Sip and freeswitch servers are on amazon ec2 each. Physical servers are different. Firstly, I test the opportunity to call to freeswitch and to receive the answer. From the part of opensips everything works, as far as I can judge, correctly. A problem appears at a call with use of freeswitch. In particular for tests I call on number 5000 through Zoiper . In reply I receive the mistake: 488 INCOMPATIBLE_DESTINATION Here is the text of freeswitch tracking data: EXECUTE sofia/external/8@SIP_DOMAIN set(rtp_secure_media=true) 2016-03-14 02:54:02.731306 [DEBUG] mod_dptools.c:1498 SET sofia/external/8@SIP_DOMAIN [rtp_secure_media]=[true] EXECUTE sofia/external/8@SIP_DOMAIN hash(insert/172.31.22.124-spymap/8/89574eb2-e9b1-11e5-97c9-1f13b1bd8007) EXECUTE sofia/external/8@SIP_DOMAINhash (insert/172.31.22.124-last_dial/8/5000) EXECUTE sofia/external/8@SIP_DOMAIN hash(insert/172.31.22.124-last_dial/global/89574eb2-e9b1-11e5-97c9-1f13b1bd8007) EXECUTE sofia/external/8@SIP_DOMAIN export(RFC2822_DATE=Mon, 14 Mar 2016 02:54:02 -0400) 2016-03-14 02:54:02.731306 [DEBUG] switch_channel.c:1293 EXPORT (export_vars) [RFC2822_DATE]=[Mon, 14 Mar 2016 02:54:02 -0400] EXECUTE sofia/external/8@SIP_DOMAIN answer() 2016-03-14 02:54:02.731306 [DEBUG] switch_core_media.c:4161 Audio Codec Compare [GSM:3:8000:20:13200:1]/[G722:9:8000:20:64000:1] 2016-03-14 02:54:02.731306 [DEBUG] switch_core_media.c:4161 Audio Codec Compare [GSM:3:8000:20:13200:1]/[PCMU:0:8000:20:64000:1] 2016-03-14 02:54:02.731306 [DEBUG] switch_core_media.c:4161 Audio Codec Compare [GSM:3:8000:20:13200:1]/[PCMA:8:8000:20:64000:1] 2016-03-14 02:54:02.731306 [DEBUG] switch_core_media.c:4161 Audio Codec Compare [GSM:3:8000:20:13200:1]/[GSM:3:8000:20:13200:1] 2016-03-14 02:54:02.731306 [DEBUG] switch_core_media.c:4216 Audio Codec Compare [GSM:3:8000:20:13200:1] ++++ is saved as a match 2016-03-14 02:54:02.731306 [WARNING] switch_core_media.c:4234 Crypto not negotiated but required. 2016-03-14 02:54:02.731306 [DEBUG] switch_core_media.c:4492 No 2833 in SDP. Disable 2833 dtmf and switch to INFO 2016-03-14 02:54:02.731306 [NOTICE] switch_channel.c:3806 Hangup sofia/external/8@SIP_DOMAIN [CS_EXECUTE] [INCOMPATIBLE_DESTINATION] 2016-03-14 02:54:02.731306 [DEBUG] switch_core_session.c:2796 sofia/external/8@SIP_DOMAIN skip receive message [APPLICATION_EXEC_COMPLETE] (channel is hungup already) 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:539 (sofia/external/8@SIP_DOMAIN) State EXECUTE going to sleep 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:473 (sofia/external/8@SIP_DOMAIN) Running State Change CS_HANGUP 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:739 (sofia/external/8@SIP_DOMAIN) Callstate Change RINGING -> HANGUP 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:741 (sofia/external/8@SIP_DOMAIN) State HANGUP 2016-03-14 02:54:02.731306 [DEBUG] mod_sofia.c:431 Channel sofia/external/8@SIP_DOMAIN hanging up, cause: INCOMPATIBLE_DESTINATION 2016-03-14 02:54:02.731306 [DEBUG] mod_sofia.c:568 Responding to INVITE with: 488 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:60 sofia/external/8@SIP_DOMAIN Standard HANGUP, cause: INCOMPATIBLE_DESTINATION 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:741 (sofia/external/8@SIP_DOMAIN) State HANGUP going to sleep 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:508 (sofia/external/8@SIP_DOMAIN) State Change CS_HANGUP -> CS_REPORTING 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:473 (sofia/external/8@SIP_DOMAIN) Running State Change CS_REPORTING 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:827 (sofia/external/8@SIP_DOMAIN) State REPORTING 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:104 sofia/external/8@SIP_DOMAIN Standard REPORTING, cause: INCOMPATIBLE_DESTINATION 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:827 (sofia/external/8@SIP_DOMAIN) State REPORTING going to sleep 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:499 (sofia/external/8@SIP_DOMAIN) State Change CS_REPORTING -> CS_DESTROY 2016-03-14 02:54:02.731306 [DEBUG] switch_core_session.c:1646 Session 1 (sofia/external/8@SIP_DOMAIN) Locked, Waiting on external entities 2016-03-14 02:54:02.731306 [NOTICE] switch_core_session.c:1664 Session 1 (sofia/external/8@SIP_DOMAIN) Ended 2016-03-14 02:54:02.731306 [NOTICE] switch_core_session.c:1668 Close Channel sofia/external/8@SIP_DOMAIN [CS_DESTROY] 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:630 (sofia/external/8@SIP_DOMAIN) Running State Change CS_DESTROY 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:640 (sofia/external/8@SIP_DOMAIN) State DESTROY 2016-03-14 02:54:02.731306 [DEBUG] mod_sofia.c:341 sofia/external/8@SIP_DOMAIN SOFIA DESTROY 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:111 sofia/external/8@SIP_DOMAIN Standard DESTROY 2016-03-14 02:54:02.731306 [DEBUG] switch_core_state_machine.c:640 (sofia/external/8@SIP_DOMAIN) State DESTROY going to sleep Error in my_thread_global_end(): 1 threads didn't exit In such a case I use the following configuration of freeswitch: vars.xmp: http://pastebin.com/DijwsQ7j modules.conf.xml: http://pastebin.com/PjeNgVkh At the same time I use the following configuration of opensips: http://pastebin.com/49XpcGBj Whats i doing incorrect?

3 6

Connecting FreeSwitch to Genesis
by Shlomi Schwartz 14 Mar '16

14 Mar '16

Sorry if this is a stupid question: I've connected to FreeSwicth using mod_verto, now I would like to route the call to a Genesys via SIP. Is it the right approach? is there any documentation for it? The purpose is to be able to integrate my webRTC solution to many contact center platforms. -- This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this on behalf of the addressee you must not use, copy, disclose or take action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply email and delete this message. Thank you.

3 5

Call failed with mod_verto
by bhavik patel 14 Mar '16

14 Mar '16

Hi All, I have Freeswitch-1.6.5 running on Debian-8.2. I want to use mod_verto for webrtc and followed steps mentioned at https://freeswitch.org/confluence/display/FREESWITCH/mod_verto. But calls are not working. Outbound call fails with NORMAL_TEMPORARY_FAILURE. When make inbound call it rings at destination and drops as soon as it answer. Here i am attaching logs for both type of calls. Let me know if you require anything else from my end. Thanks

3 9

FS behind NAT
by Tao Zhang 14 Mar '16

14 Mar '16

Hi all, I have been tracking down why my FS verto video conf demo setup not working. Now I believe the reason is my FS server is behind a NAT. My previous posts may be misleading now, so I am creating a new thread. Some evidence I saw: 1) In the SDP dialog, I saw my FS server's internal IP got used, rather than its public IP, for example Dialog 6c62f631-469c-a2f0-e484-6e9cfcda7eedANSWER SDP v=0 o=FreeSWITCH 1444254094 1444254095 IN IP4 10.240.15.244 s=FreeSWITCH c=IN IP4 10.240.15.244 <============= t=0 0 2) Also, in wireshark , I saw my client is trying to send STUN packets to the FS server's internal IP, 450 18.252726000 192.168.1.7 10.240.15.244 STUN 158 Binding Request user: mlvLvmDzCphyQYz5:N3MZoB99aGiAZotr So now my questions are Will the verto video conf demo work when the FS server is behind NAT? If so, could someone help me with the necessary config changes and etc.? Thanks so much! Tao

3 6