August 2010 - FreeSWITCH-users - FreeSWITCH Mailing Lists

Voice mail isn't working on one extension.
by Jim 27 Aug '10

27 Aug '10

I have the following connected to my FS box. A GrandStream phone, two POTS phones via the Linksys SPA2002, a soft phone installed on another system, not the FS machine and another POTS connected via an A200 port. If I call the GrandStream from one of the POTS on the SPA2002, it rings for ~30 seconds and I get a fast busy. If I call on the A200 connected phone, it rings and then nothing. It just stops ringing. If I call on the soft phone after it stops ringing I see a 404 error displayed on the phone "console". The log shows: 2010-08-26 06:30:59.364714 [DEBUG] mod_sofia.c:453 Channel sofia/internal/1000(a)192.168.2.51 hanging up, cause: NO_ROUTE_DESTINATION 2010-08-26 06:30:59.384660 [DEBUG] mod_sofia.c:515 Responding to INVITE with: 404 2010-08-26 06:30:59.384660 [DEBUG] switch_core_state_machine.c:46 sofia/internal/1000(a)192.168.2.51 Standard HANGUP, cause: NO_ROUTE_DESTINATION If I do a diff of directory/default/1001.xml and 1002.xml the only differences are the numbers and passwords. Any idea what I might want to try next? Thanks, Jim.

3 10

Deployment questions: Freeswitch or Freeswitch with a SIP proxy server
by Kamigishi Rei 27 Aug '10

27 Aug '10

Hello, I need to set up a relatively small (up to, probably, 5000 users at maximum) SIP service; however I'm not sure if what I need can be implemented by using Freeswitch, so I decided to consult this mailing list. First, the essential part: local calls within the SIP realm; easy to get implemented and basically works out of the box with some scripts to export data from our billing database and to convert that data into XML. The second task, implementing a PSTN gateway, looks pretty straightforward, too. However, there are a few questions about something else: external SIP servers/operators, and DNS/ports setup. Basically, I can receive inbound calls from other SIP operators to my public extensions out of the box, without setting up any gateways and such; that behaviour is welcome. However, the invite domain presented gets changed to FreeSwitch's IPv4 address before showing up on the softphone. Is overriding it with "{sip_invite_domain=${sip_from_host}}sofia/internal/$1%$${domain}" in a bridge application considered bad practice? Placing outbound calls to other SIP operators: this "sort of" works if I uncomment   in dialplan/default.xml. However, this is a very weird and - to me - extremely crude solution as, first, it's a fallback rule (thus we can't call a remote extension that also exists in our system - f.ex. calling 1001@remote will result in the call being directed to 1001@local), and second, ${domain_exists(${sip_req_host})} is a really lousy check and doesn't account for DNS aliases, IP addresses, etc. thus resulting in possible fallback to a direct call instead of a dialplan route if the sip_invite_domain wasn't set properly and we got a call back. It's even more confusing if we use DNS NAPTR and SRV records. Currently, I have the following in my dialplan: <extension name="external_domain_outbound_calls" continue="true"> <condition field="source" expression="mod_sofia"/> <condition field="${outside_call}" expression="^$"/> <condition field="${sip_to_host}" expression="^((${domain}|sip.${domain})|${local_ip_v4}|)$"> <anti-action application="bridge" data="{sip_invite_domain=$${domain}}sofia/${use_profile}/${sip_to_uri}"/> </condition> </extension> Does it seem acceptable or is that bad practice as well? Is placing outbound calls to external SIP URIs considered bad practice overall when using Freeswitch? From what I've gathered @ the irc channel, "Freeswitch is not a proxy", so should I instead use something like OpenSIPS for such tasks? And about the DNS settings: It is established that the internal profile serves authenticated clients, and the external profile is for untrusted connections (gateways). I presume that to receive calls to my public extensions, I need the remote SIP operators/servers/clients to be able to find my "public" SIP port, which is 5080 in the default configuration. Therefore, I have the IN SRV records for "mydomain.org" pointing at port 5080, external profile. Or is it acceptable to receive external calls to public extensions to the internal profile? (Clarification needed here; is "external" profile supposed to be "for external gateways", or "for any untrusted connections, including calls from non-gateways to our public extensions". Also, is accepting calls from a non-authenticated SIP peer to the public extensions, but on the internal profile's port, not recommended?) However because of that I have to use a DNS alias (or just the IP address) in the clients' (f.ex. softphones) configuration; f.ex. "sip.mydomain.org". This results in clients using "sip.mydomain.org" as their authentication realm (which is understandable), and direct - non-proxied - outbound calls appear as "ext(a)sip.mydomain.org" on the remote side. Which leads to calls originating from the remote side based on their call history arriving on port 5060, internal profile. If I specify my Freeswitch server as the SIP proxy, the aforementioned external_domain_outbound_calls extension takes care of it by replacing sip_invite_domain in the headers. But since "Freeswitch is not a proxy", I guess there should be some other solution instead of having to place calls to external URIs via Freeswitch? Overall, if accepting unauthenticated inbound calls on the internal profile is fine and they still go into public context, I guess it's fine to have the DNS point at the internal profile? (Since gateways can connect to the external profile's port explicitly.) Thanks in advance. -- Kamigishi Rei KREI-RIPE

2 2

Re: [Freeswitch-users] any disaster lurking in this over all plan?
by Robust Process 27 Aug '10

27 Aug '10

>The only disaster I see is the Aastra in the mix. Thats just my personal preference. thanks for the response Brian, since I've never had my hands on any of the IP phones, what feature, lack of feature, quality issue, or configuration problems have you had with the Aastra's compared to other IP phones? Orin

2 1

FreeSWITCH Sofia DNS Cache SIP/SDP INVITE Auth Fail
by Locutis of Borg 27 Aug '10

27 Aug '10

Hi, I have been struggling with failed outbound calls to my ITSP. Seems that my FreeSWITCH does not know who to send SIP/SDP INVITE packets to. FS has the DNS cached and uses that to send the first INVITE, then it gets expired and refreshes it to another IP, just in time for the second INVITE, causing an AUTHENTICATION failure. The packet capture would only show the 1st INVITE SIP to one resolved IP. Then, the 2nd INVITE SIP/SDP to a different resolved IP. FS is on a DMZ. I have extensively tested IPTABLES, NAT, and even rebuilt FS from the latest GIT 2 days ago. In the ext. SIP profile, if I use hostname for proxy, initial calls fail, but on redial, they complete. Now, in the same SIP profile, I use the IP address, call are OK. (for both, REG is false) A reasonable solution seems to be to use the IP in the ext. SIP profile. But, is that the best? Any experience with this kind of issue? Any suggestion would be greatly appreciated. Thank you, Cheers

4 4

issue with media on two internal gateways
by Michael Scheidell 26 Aug '10

26 Aug '10

I don't want FS1 who talks to FS2 to try to use the PUBLIC rtp port. but I can't seem to see how. FS1: 192.168.0.3 (public ip 204.89.241.135) listeing for trunk calls on 5060 FS2: 192.168.0.2 (public ip 204.89.241.150) listening for trunk calls on 5080 FS1 -> FS2 in Inivite: (then it tries to bind media to public ip, i get one way audio, and no DTMF) Contact: <sip:gw+secnap.com@204.89.241.135:5060;transport=udp> Contact Binding: <sip:gw+secnap.com@204.89.241.135:5060;transport=udp> URI: <sip:gw+secnap.com@204.89.241.135:5060;transport=udp> SIP contact address: sip:gw+secnap.com@204.89.241.135:5060 (204.89.241.135 is the public ip, but these two boxes are internal) FS2 listens on port 5080, FS1 listens on port 5060. FS2-> FS1 (what is fair is FAIR..) Contact: <sip:gw+secnap.com@204.89.241.150:5080;transport=udp> Contact Binding: <sip:gw+secnap.com@204.89.241.150:5080;transport=udp> URI: <sip:gw+secnap.com@204.89.241.150:5080;transport=udp> SIP contact address: sip:gw+secnap.com@204.89.241.150:5080 if tried so far: put outbound-proxy in secnap.com.xml <param name="rtp-ip" value="192.168.0.3"/> <param name="sip-ip" value="192.168.0.3"/> in sofia.conf.xml (noop) in acl.conf.xml: <node type="allow" cidr="192.168.0.2/32"/> no.. internal xml: --> <param name="ext-rtp-ip" value="auto-nat"/> <param name="ext-sip-ip" value="auto-nat"/> --> <param name="ext-rtp-ip" value="auto"/> <param name="ext-sip-ip" value="auto"/> in external.xml also making lots of calls from my cell phone.. watching traffic. -- Michael Scheidell, CTO o: 561-999-5000 d: 561-948-2259 ISN: 1259*1300 > *| *SECNAP Network Security Corporation * Certified SNORT Integrator * 2008-9 Hot Company Award Winner, World Executive Alliance * Five-Star Partner Program 2009, VARBusiness * Best in Email Security,2010: Network Products Guide * King of Spam Filters, SC Magazine 2008 ______________________________________________________________________ This email has been scanned and certified safe by SpammerTrap(r). For Information please see http://www.secnap.com/products/spammertrap/ ______________________________________________________________________

2 4

Enterprise setup registration question
by Henry Huang 26 Aug '10

26 Aug '10

Hi: I have followed wiki: http://wiki.freeswitch.org/wiki/Enterprise_deployment_OpenSIPS To deploy OpenSIPS in front of FreeSWITCH. I have played with the load balancer part before, so I know how that works. I have now been able to register through the OpenSIPS to FreeSWITCH box via the dispatcher module of OpenSIPS. But what happened is that the registration IP is the IP of the OpenSIPS server, not the end user's IP address. I am wondering how would we process incoming call for the registered user? Obviously the call will be routed back to the OpenSIPS server with an invite, which gets processed by the load balance module according to the script. So what I am asking is that is this a logic flaw in the original script? or did I misunderstood it? If it is a flaw, does any one know how to fix the senario? Because I am not that good with OpenSIPS to fix it. Thanks, Henry Huang Unified Communication System R&D US: +1 (626) 606-3306 台灣(Taiwan): +886 933847619 Chat Google Talk: red_rain_seven(a)gmail.com Skype: unicsolution MSN: b_ball_henry(a)hotmail.com Contact Me [image: Linkedin]<http://www.linkedin.com/pub/henry-huang/1/654/578>[image: Facebook] <http://www.facebook.com/profile.php?id=595148342>[image: Twitter]<http://twitter.com/unicsolution>

1 0

Misc. Dialplan Tools say types
by David Swardstrom 26 Aug '10

26 Aug '10

I tried using the type "TELEPHONE_NUMBER" and got a failure message: <command> session.execute("say", "en TELEPHONE_NUMBER pronounced " + myani]); </command> <error> mod_say_en.c:481 Unknown Say type=[9] </error> There appears to be no code to support TELEPHONE_NUMBER in mod_say_en.c. Note: I looked. I have not done a code refresh since before ClueCon. Has it been added or is this part of code from another language. Actually, it seems that it may be in French (mod_say_fr.c). -- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Misc-Dialplan-Tools-say-types… Sent from the freeswitch-users mailing list archive at Nabble.com.

2 1

Using streamFile to steam from memory
by David Swardstrom 26 Aug '10

26 Aug '10

I would like to be able to extract a short voice segment from a Database and stream it directly from memory using JavaScript. Is there a version of streamFile which would do this? -- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Using-streamFile-to-steam-fro… Sent from the freeswitch-users mailing list archive at Nabble.com.

2 1

what is manual_calls in fifo
by Abubacker siddiq 26 Aug '10

26 Aug '10

Dear community , whenever I issue the command "fifo list" it shows all the fifo names along with the manual_calls actually what it is , Is it a default fifo in the freeswitch ? . and also why the status for a member always nothing even he is busy with the customer ,Is this also a bug ? $ fifo list <fifo_report> <fifo name="test_fifo" consumer_count="0" caller_count="0" waiting_count="0" importance="0" outbound_per_cycle="1" outbound_priority="5" outbound_strategy="ringall"> <outbound> <member simo="1" use_count="0" timeout="60" lag="1" outbound-call-count="0" outbound-fail-count="4" taking-calls="1" status=""outbound-call-total-count="0" outbound-fail-total-count="4" logged-on-since="2010-08-26 11:57:46" manual-calls-out-count="0" manual-calls-in-count="0" manual-calls-out-total-count="0" manual-calls-in-total-count="0" ring-count="0" start-time="never" stop-time="never" next-available="2010-08-26 12:26:23">{member_wait=nowait}user/1000</member> </outbound> <callers></callers> <consumers></consumers> <bridges></bridges> </fifo> <fifo name=*"manual_calls*" consumer_count="0" caller_count="0" waiting_count="0" importance="0" outbound_per_cycle="0" outbound_priority="5" outbound_strategy="ringall"> <outbound> <member simo="1" use_count="0" timeout="60" lag="1" outbound-call-count="0" outbound-fail-count="4" taking-calls="1" status="" outbound-call-total-count="0" outbound-fail-total-count="4" logged-on-since="2010-08-26 11:57:46" manual-calls-out-count="0" manual-calls-in-count="0" manual-calls-out-total-count="0" manual-calls-in-total-count="0" next-available="2010-08-26 12:26:23">{member_wait=nowait}user/1000</member> </outbound> <callers></callers> <consumers></consumers> <bridges></bridges> </fifo> </fifo_report> Thanks in advance !!! -- *Best Regards, **Abubacker systems engineer bk systems (p) ltd** *

3 4

Cause IE in Progress Message (Sangoma, LibPRI)
by sdemers＠q4innovation.com 26 Aug '10

26 Aug '10

Hi everyone, I'm trying to catch the CAUSE IE into the ISDN NI2 Progress Message with Sangoma A108 card using LibPRI. Based on the Log bellow, the OZMOD_libpri catch the CAUSE IE, but at the event_socket, there is no element with that specification. This is very important because the most carrier doesn't send (90% of the time) the SIT tone UNALLOCATED, and also return NORMAL_CLEARING cause in the disconnect. They send a progress message with an Unallocated (unassigned) Cause IE and play a message on the voice path for a end user. When you have a machine on the other side it's not easy to catch. I've take a look into the ozmod_libpri.c and I didn't found any code related to LPWRAP_PRI_EVENT_PROGRESS. The print-out come probably from LPWRAP_PRI_EVENT_ANY (on_anything) function. Is there a way to push it to the event_socket remote application? Thanks for your answer. ---------------------------------------------------------------------- 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 Handling message for SAPI/TEI=0/0 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 -- ACKing all packets from 4 to (but not including) 5 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 -- Since there was nothing left, stopping T200 counter 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 -- Stopping T203 counter since we got an ACK 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 -- Nothing left, starting T203 counter 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < Protocol Discriminator: Q.931 (8) len=13 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < Call Ref: len= 2 (reference 3/0x3) (Terminator) 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < Message type: PROGRESS (3) 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < [08 02 80 81] *2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < Cause (len= 4) [ Ext: 1 Coding: CCITT (ITU) standard (0) Spare: 0 Location: User (0)* *2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < Ext: 1 Cause: Unallocated (unassigned) number (1), class = Normal Event (0) ]* 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < [1e 02 80 88] 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < Progress Indicator (len= 4) [ Ext: 1 Coding: CCITT (ITU) standard (0) 0: 0 Location: User (0) 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 < Ext: 1 Progress Description: Inband information or appropriate pattern now available. (8) ] *2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 -- Processing IE 8 (cs0, Cause)* 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 -- Processing IE 30 (cs0, Progress Indicator) 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 Sending Receiver Ready (6) 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 > [ 02 01 01 0c ] 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 > Supervisory frame: 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 > SAPI: 00 C/R: 1 EA: 0 > TEI: 000 EA: 1 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 > Zero: 0 S: 0 01: 1 [ RR (receive ready) ] > N(R): 006 P/F: 0 > 0 bytes of data 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:106 -- Restarting T203 timer 2010-08-26 11:01:08.849421 [DEBUG] ozmod_libpri.c:979 Caught Event span 1 17 (PROGRESS)

1 0