- FreeSWITCH-users - FreeSWITCH Mailing Lists

Call to unregister endpoint with auth=true after endpoint password change
by Jose David Jurado Alonso 25 Jul '19

25 Jul '19

Hi, I have FS configured to only send calls to authenticated endpoints. It works very well in general but there is a case in which it sends the call to unregistered endpoints: 1. Auth the endpoint and "registration" (ok) 2. Call to endpoint and call arrives (ok) 3. Change the endpoint password in FS DB and endpoint "fail registration" 4. Call to endpoint and call arrives (ERROR, call must be ignored) The ERROR is because the REGISTER event know that endpoint "fail registration" and yet send the call anyway ignoring the "auth_calls=true" configuration. If I execute these fs_cli commands between step 3 and 4, the call is not sent (OK for me, expected and desired case): > sofia profile external flush_inbound_reg all > sofia profile external flush_inbound_reg reboot > sofia profile external check_sync I have not tested if all the commands are necessary or not. Does anyone know how it can be done to make it work as desired using configuration without having to execute the commands after every password change in an endpoint? Relevant configuration is: vars.xml:  <X-PRE-PROCESS cmd="set" data="internal_auth_calls=true"/> <X-PRE-PROCESS cmd="set" data="internal_sip_port=5060"/> <X-PRE-PROCESS cmd="set" data="internal_tls_port=5061"/> <X-PRE-PROCESS cmd="set" data="internal_ssl_enable=false"/>  <X-PRE-PROCESS cmd="set" data="external_auth_calls=true"/> <X-PRE-PROCESS cmd="set" data="external_sip_port=5080"/> <X-PRE-PROCESS cmd="set" data="external_tls_port=5081"/> <X-PRE-PROCESS cmd="set" data="external_ssl_enable=false"/> sip_profile "external.xml" (i use it): <param name="manage-presence" value="false"/> <param name="auth-calls" value="true"/> <param name="multiple-registrations" value="true"/> I use a dynamic directory getting SIP accounts from DB and generate the XML (mod_curl) setting "cacheable=0". Thanks! José David Jurado Alonso.

4 4

Missing remote sound in recordings using OPUS codec
by Allan Kristensen 23 Jul '19

23 Jul '19

Hello, I'm missing remote sound when recording call using the Opus codec. The sound works perfectly during the call, but when listing to recording afterwards, remote sound is missing (sound from SIP UA / A-leg). When using PCMA/PCMU codec recording works fine. *To use OPUS I do:* Set 'absolute_codec_string', 'OPUS@8000h@20i@1c' *To start the recordin i do:* <action application="record_session" data="$${RECORDING_DIR}/${uuid}.wav"/> Extra info: The call is being transcoded to PCMA/PCMU to carrier when using Opus codec. Tried both v1.8 branch and master and using Opus lib 1.3-1 Does transcoding cause the problems perhaps? or is it mod_opus? Thank you... /Allan

2 1

Corrupted core.db
by Jai Rangi 23 Jul '19

23 Jul '19

We are getting this error When we do show registration switch_core_sqldb.c:1197 SQL ERR: [select * from registrations where hostname='freeswitch'] database disk image is malformed If there a way to fix this without restarting free-switch. Thank you -Jai

11 20

Call Center sound announcement
by Moshe Rosenberg 23 Jul '19

23 Jul '19

Hi i just came acorss this bug report https://freeswitch.org/jira/browse/FS-10253 while i was searching to fix my issue i see this is a few years old. did anyone find a way around this bug? -- Moshe Rosenberg

3 3

How to configure External Gateway with Multiple NICs
by Vishal Kakkar 22 Jul '19

22 Jul '19

Hi all, My freeswitch server has two NICs eth0 and eth1. eth0 - Connected to internet for public/remote SIP Gateways eth1 - Connected to ISP(non internet) SIP server directly. Now to have all external gateways working together i need to specify which sip-ip, rtp-ip to be used in each gateway. But i can see these settings can be done only in external.xml So if i specify IP of any eth0 or eth1, all external gateways (be public or local ISP) registrations go using VIA tag of that very IP address, which is causing failed registrations. If i give eth0 IP then public SIP gateways work, but not local SIP SIP. If i give eth1 IP then local ISP SIP gateway work, but not public/remote gateways. How can such situations be handled. Is there any way to specify via header/sip,rtp-ip per gateway. Thanks Vishal.

2 1

Problem with Dial Plan and Diversion Headers
by Rick Ewart 22 Jul '19

22 Jul '19

Hi all. I am having an issue with calls not being routed correctly in our freeswitch system and I am having trouble figuring out the real cause. We are using the following in our routing plan to determine the number being called on an incoming call, which seems pretty straight forward: <condition break="never"> <action inline="true" application="set" data="dest_num=${destination_number}"/> <action inline="true" application="set" data="did_num=0"/> </condition> Generally speaking it works fine - we have been in production for years now doing many millions of calls. But an issue has cropped up with a call that comes in and is not being routed properly. Below is the SIP INVITE and header we receive (sanitized to protect the guilty). You will note that the INVITE has 15615551234 which is the correct DID on our system being called. However the "To" header specifies 14145241234 which if I understand what I read, is apparently what the caller dialed and it was diverted to our DID. Since my system never heard of 14145241234, it routed it to our default "Not in Service" message. In discussing with my carrier, they are saying its being delivered correctly and my system must be setup improperly for this situation. Since I never heard of the diversion header before this, I am at a bit of a loss (even after reading some articles on it). I thought I had a pretty good handle on this stuff but this is throwing me for a loop as it seems to me that the "To" should have been rewritten on the far end B2BUA (or somewhere before it got to me) to be the same as the INVITE but apparently not? I am trying to figure out what would be the most failsafe variable to use if destination_number is not it (as its what I see in all the FS Doc examples and elsewhere). Seems the INVITE is the only place the real number exists in this situation. I am searching for what that variable would be correct as the Channel Variables docs don't contain a lot of descriptions. I am thinking perhaps it should be sip_req_user? Since it's a production box, I figured I should ask the experts rather than guess on my own and make people grumpy due to some other unintended consequence I don't realize. I appreciate any help/explanation you can/are willing to provide. Thanks! Rick Ewart Session Initiation Protocol (INVITE) Request-Line: INVITE sip:15615551234@123.123.41.36:5060 SIP/2.0 Message Header Via: SIP/2.0/UDP 68.232.9.155:5060;branch=z9hG4bK0cB87133736bc4b79bd From: "ABC COMPANY" <sip:15615554162@68.232.9.155>;tag=gK0c5693cd To: <sip:14145241234@123.123.41.36> Call-ID: 191660912_117119432(a)68.232.9.155 CSeq: 271179 INVITE Max-Forwards: 70 Allow: INVITE,ACK,CANCEL,BYE,OPTIONS Accept: application/sdp Contact: "ABC COMPANY" <sip:15615554162@68.232.9.155:5060> Remote-Party-ID: "ABC COMPANY" <sip:15615554162@68.232.9.155:5060>;privacy=off Diversion: <sip:+1414524123@68.232.9.155:5060>;privacy=off;screen=no;reason=unknown;c ounter=1 Supported: replaces Content-Length: 281 Content-Disposition: session; handling=required Content-Type: application/sdp

2 1

Re: [Freeswitch-users] freeswitch package on ARM (raspberry pi)
by Igor Olhovskiy 19 Jul '19

19 Jul '19

There is a repo for arm maintaned by FusionPBX. echo "deb https://repo.fusionpbx.com/armhf stretch 1.8.4" > /etc/apt/sources.list.d/freeswitch.list echo "deb https://repo.fusionpbx.com/armhf jessie 1.6.20" > /etc/apt/sources.list.d/freeswitch.list curl https://repo.fusionpbx.com/public.key | apt-key add - On Jun 26 2019, at 8:22 pm, Bipin Patel via FreeSWITCH-users <freeswitch-users(a)lists.freeswitch.org> wrote: > hi, > > i was wondering if FS packages will be available for ARM raspbian > because we have been running multiple SOHO PBX using rpi for clients for > quiet some time now, currently we build from master on raspbian stretch > as well as jessie and it works well. We did some stress test with 50 > calls and 100 channels using rpi 3b+ with g711u and it works flawless as > well as we have it on rpi 2 also easily able to handle 20+ calls and now > with the rpi 4 im sure it can handle a lot more easily so why not > officially create packages for ARM rather than have to compile from > master. We also developed a custom CDR portal we supply with it which > gives the user the ability to create extenions, gateways etc etc as xml > rather than use a DB or xml curl etc to keep the resource consumption > low. > > > Regards, > Bipin > > > _________________________________________________________________________ > The FreeSWITCH project is sponsored by SignalWire https://signalwire.com > Enhance your FreeSWITCH install with disruptive priced SMS and PSTN services. > Build your next product on our scalable cloud platform. > > Join our online community to chat in real time https://signalwire.community > Professional FreeSWITCH Services > sales(a)freeswitch.com > https://freeswitch.com > > Official FreeSWITCH Sites > https://freeswitch.com/oss > https://freeswitch.org/confluence > https://cluecon.com > > 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 > https://freeswitch.com >

4 3

Mod_Python
by Caio Assis 19 Jul '19

19 Jul '19

Good afternoon. I use mod_python to execute scripts, and I notice that when there's a relative high call traffic, freeswitch's performance decreases significantly. And I also noticed that, the more interaction the script has with freeswitch's session object (getVariable, execute, etc) the more the performance decreases. I kept the interaction to the minimum, as in 1 session.execute('multiset'). Any sugestions?

2 1

strange behavior when bridge fails.
by Caragea Silviu 18 Jul '19

18 Jul '19

Hello, I'm using last stable FS 1.8.7 and when I'm doing a bridge to a not existing destination I can see that even after bridge fails and the LEG a receives 480, FS is still re-transmitting the INVITE few times. Before bridge I'm setting the following vars: continue_on_fail=true, hangup_after_bridge=true, progress_timeout=5, bypass_media=true And then I'm bridging using: {origination_caller_id_name=400000000,origination_caller_id_number=400000000,ignore_early_media=consume}sofia/external/ 44444444@172.17.33.28:6060 Bridge fails with PROGRESS_TIMEOUT after 5 seconds which is correct but on PCAP I can still see lot of retransmission's for the INVITE. Please see the attached image as well. 172.17.33.27 is the caller, 172.17.33.111 is FS and 172.17.33.28 is the IP where we are doing the bridge and where no server listen. It's this expected behavior ? Even after 16 seconds after the bridge fails the reinvite is sent. [image: Screen Shot 2019-07-18 at 12.46.53 PM.png] Silviu

1 0

Limit cps per source IP and port
by António Silva 18 Jul '19

18 Jul '19

Hi, is it possible to limit the number of calls per source ip and port before the dialplan? I know that this can be done using mod_limit in the dialplan, i was thinking of a limit like the sessions-per-second global limit, this way less resources are used if some ip/port is flooding our machine. Another possibility is using the firewall... but i like this option less because i wont know of the drop calls. Thanks, -- Saludos / Regards / Cumprimentos António Silva

1 0