December 2009 - freeswitch-dev@lists.freeswitch.org

Re: [Freeswitch-dev] mod event multicast drops some events on outbound call legs ?
by Matteo 09 Dec '09

09 Dec '09

Hi again, I've found a fix, but maybe the error is hidded somewhere else. If in the function SWITCH_MODULE_RUNTIME_FUNCTION(mod_event_multicast_runtime), line 515 in mod_event_multicast.c I change switch_yield(100000); to a lower time, for example switch_yield(1000); it works ok. so maybe a loo long sleep time makes the multicast receiver loose packets if we have many events (I'm sending all the events for now, maybe with filters this will not happen). So, with switch_yield(1000); works ok. But this's the right fix? regards, matteo

2 2

[Freeswitch-dev] FreeSWITCH 1.0.5 is (almost) here!
by Michael Collins 08 Dec '09

08 Dec '09

Greetings, The FreeSWITCH developers have uploaded the latest and greatest FreeSWITCH 1.0.5 pre-release version. Please check out the release announcement<http://www.freeswitch.org/node/220>. Let's all get updated as soon as possible. Also, please report bugs right away and follow up when the developers need further information. We have had to close out some bugs due to lack of information from the one reporting. Of course, those running SVN trunk are asked to do a "make current" as soon as reasonably possible. The devs love it when you are on the latest trunk. :) Thanks again for all of your help! Let's keep up the good work and we'll have 1.0.5 available in no time. -Michael

1 0

[Freeswitch-dev] Aastra XML scripts.
by Reece Savage 04 Dec '09

04 Dec '09

Would anyone be willing to port the Aastra XML scripts for Asterisk to FreeSWITCH? I would be willing to sponser. Reece Savage Information Technology Manager King & Ballow Law Offices 315 Union Street Suite 1100 Nashville, TN 37201 Phone (615) 726-5525 Fax (615) 254-7907 rsavage(a)kingballow.com <mailto:rsavage@kingballow.com>

1 0

[Freeswitch-dev] FreeSWITCH Weekly Conference Call Starting!
by Michael Collins 04 Dec '09

04 Dec '09

FYI, The agenda is here: http://wiki.freeswitch.org/wiki/FS_weekly_2009_12_04 Please call in! :) -Michael

1 0

[Freeswitch-dev] uuid_bridge kills both channels if they are executing java app
by Artem Shiyanov 04 Dec '09

04 Dec '09

Hi there! This message is a forward from user-mail-list. I'm trying to fix such a problem: FreSwithch compiled from SVN-trunk, date = 11/02/2009. What is need: connect two users, initially one is on the home-grown java-based IVR and other party is off hook. What is done/got: User1 is on the java application, it represents simple IVR system, and the most used FS API operation is "streamFile". User2 is off hook. next: (mod_socket) create_uuid <get uuid_x> bgapi originate {origination_caller_id_name=User1}[origination_uuid=uuid_x]User1 &park() <User1 answers incomming call> <get event channel_User1 answered> <get event channel_User1 parked> uuid_bridge uuid_User1 uuid_User2 <get event channel_User1 hangup, cause=NORMAL_CLEARING> <get event channel_User2 hangup, cause=*DESTINATION_OUT_OF_ORDER*> <both channel_User1 and channel_User2 are down> FS log is here: http://pastebin.freeswitch.org/11380 Thank you much for any help, Artem ---------- Forwarded message ---------- From: Anthony Minessale <anthony.minessale(a)gmail.com> Date: Wed, Dec 2, 2009 at 10:24 PM Subject: Re: [Freeswitch-users] uuid_bridge kills both channels if they are executing java app To: freeswitch-users(a)lists.freeswitch.org you should be working on SVN trunk if you are doing development, we are so far forward from 1.0.4 we can't do debugging very easily. I don't know all of the details of what you are trying to do but you are hitting some race conditions because of the async nature of the socket connection and the way you are using it. On Wed, Dec 2, 2009 at 1:08 PM, Artem Shiyanov <shiyanov(a)gmail.com> wrote: > I'm back again with the same issue. > Now it is became worse: it reproduces occasionally. > [FS version is 1.04, test_load = 2 active calls] > > I've got 2 logs: successful and not. > Here is a bad_case: > > 2009-12-02 13:27:55.159931 [NOTICE] switch_core_session.c:1576 Execute > java(/usr/local/freeswitch/scripts/fs2agi.jar > org.starpound.fs2agi.Translator > ${agi_url}) > Dec 2, 2009 1:27:55 PM org.starpound.fs2agi.Translator run > INFO: *************************************************** > Dec 2, 2009 1:27:55 PM org.starpound.fs2agi.Translator run > INFO: Run AGI application agi://localhost:4573/hello.agi?callId=929 for > session > 2898ad41-4ec1-4628-89fd-651a93a7221d > 2009-12-02 13:27:55.169841 [NOTICE] switch_cpp.cpp:1130 Run AGI application > agi://localhost:4573/hello.agi?callId=929 > 2009-12-02 13:28:02.888831 [CRIT] mod_local_stream.c:234 Leaking stream > handle! > > [switch_ivr_play_file() src/switch_ivr_play_say.c:1026] > 2009-12-02 13:28:04.799806 [NOTICE] switch_channel.c:602 New Channel > sofia/internal/2001 [76d2c0e9-16a4-4098-92c0-5977cb482e17] > 2009-12-02 13:28:05.148834 [NOTICE] sofia.c:3353 Ring-Ready > sofia/internal/2001! > 2009-12-02 13:28:05.855093 [NOTICE] sofia.c:3794 Channel > [sofia/internal/2001] has > been answered > Dec 2, 2009 1:28:05 PM org.starpound.fs2agi.Translator tellAllWeCrashed > INFO: AGI application agi://localhost:4573/hello.agi?callId=929 for session > 2898ad41-4ec1-4628-89fd-651a93a7221d crashed. Exception is: > java.lang.Exception: Internal FreeSwitch failure while streamming file, see > FreeSwitch logs for details > at > > org.starpound.fs2agi.agicommands.StreamFileCommand.execute(StreamFileCommand.java:36) > at org.starpound.fs2agi.AgiConnection.run(AgiConnection.java:48) > at org.starpound.fs2agi.Translator.run(Translator.java:56) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at > > sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39) > at > > sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) > at java.lang.reflect.Method.invoke(Method.java:597) > at org.freeswitch.Launcher.launch(Launcher.java:80) > 2009-12-02 13:28:05.870185 [NOTICE] switch_core_state_machine.c:179 Hangup > sofia/internal/2001 [CS_EXECUTE] [NORMAL_CLEARING] > 2009-12-02 13:28:05.878807 [INFO] switch_cpp.cpp:1130 AGI application > agi://localhost:4573/hello.agi?callId=929 crashed. See FS2AGI log for > details. > 2009-12-02 13:28:05.894422 [NOTICE] switch_ivr_bridge.c:667 Hangup > sofia/external/6786081291(a)66.19.38.143 [CS_SOFT_EXECUTE] > [DESTINATION_OUT_OF_ORDER] > 2009-12-02 13:28:05.894422 [NOTICE] switch_core_session.c:1086 Session 17 > (sofia/external/6786081291(a)66.19.38.143) Ended > 2009-12-02 13:28:05.894422 [NOTICE] switch_core_session.c:1088 Close > Channel > sofia/external/6786081291(a)66.19.38.143 [CS_DESTROY] > 2009-12-02 13:28:05.894422 [NOTICE] switch_core_session.c:1086 Session 18 > (sofia/internal/2001) Ended > 2009-12-02 13:28:05.894422 [NOTICE] switch_core_session.c:1088 Close > Channel > sofia/internal/2001 [CS_DESTROY] > > > > Message > "Dec 2, 2009 1:28:05 PM org.starpound.fs2agi.Translator tellAllWeCrashed > INFO: AGI application agi://localhost:4573/hello.agi?callId=929 for session > 2898ad41-4ec1-4628-89fd-651a93a7221d crashed. Exception is: > ..." > is sent from my app upon the onHangup().` > > And here is a good_case: > > 2009-12-02 13:31:45.959813 [NOTICE] switch_core_session.c:1576 Execute > java(/usr/local/freeswitch/scripts/fs2agi.jar > org.starpound.fs2agi.Translator > ${agi_url}) > Dec 2, 2009 1:31:45 PM org.starpound.fs2agi.Translator run > INFO: *************************************************** > Dec 2, 2009 1:31:45 PM org.starpound.fs2agi.Translator run > INFO: Run AGI application agi://localhost:4573/hello.agi?callId=932 for > session > 7c37369b-ffb2-4436-9288-a640047d0e5e > 2009-12-02 13:31:45.965814 [NOTICE] switch_cpp.cpp:1130 Run AGI application > agi://localhost:4573/hello.agi?callId=932 > 2009-12-02 13:31:53.648915 [CRIT] mod_local_stream.c:234 Leaking stream > handle! > > [switch_ivr_play_file() src/switch_ivr_play_say.c:1026] > 2009-12-02 13:31:59.260797 [NOTICE] switch_channel.c:602 New Channel > sofia/internal/2001 [7db554a6-861e-4492-a87b-78b6dfec6488] > 2009-12-02 13:31:59.624818 [NOTICE] sofia.c:3353 Ring-Ready > sofia/internal/2001! > 2009-12-02 13:32:00.130814 [NOTICE] sofia.c:3794 Channel > [sofia/internal/2001] has > been answered > Dec 2, 2009 1:32:00 PM org.starpound.fs2agi.Translator tellAllWeCrashed > INFO: AGI application agi://localhost:4573/hello.agi?callId=932 for session > 7c37369b-ffb2-4436-9288-a640047d0e5e crashed. Exception is: > java.lang.Exception: Internal FreeSwitch failure while streamming file, see > FreeSwitch logs for details > at > > org.starpound.fs2agi.agicommands.StreamFileCommand.execute(StreamFileCommand.java:36) > at org.starpound.fs2agi.AgiConnection.run(AgiConnection.java:48) > at org.starpound.fs2agi.Translator.run(Translator.java:56) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at > > sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39) > at > > sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) > at java.lang.reflect.Method.invoke(Method.java:597) > at org.freeswitch.Launcher.launch(Launcher.java:80) > 2009-12-02 13:32:00.149080 [INFO] switch_cpp.cpp:1130 AGI application > agi://localhost:4573/hello.agi?callId=932 crashed. See FS2AGI log for > details. > 2009-12-02 13:32:00.388838 [INFO] switch_rtp.c:1869 Auto Changing port from > 172.26.10.39:26402 to 91.190.120.190:26402 > > > > Suggestions? > > > > > > > > > > > > On Sat, Nov 21, 2009 at 12:58 PM, Artem Shiyanov <shiyanov(a)gmail.com>wrote: > >> Anthony, >> >> >>As soon as you call uuid_bridge you are transferring both legs of the >> call to bridge to each other. >> >>This means your java app must exit so the channels can connect to each >> other. >> >> I didn't know that. Now my java app is exiting upon the onHangup() call so >> everything has become "ok". Thank you much. >> I'll add note to the wiki about this issue. >> >> Artem >> >> >> >> >> On Fri, Nov 20, 2009 at 5:49 AM, Anthony Minessale < >> anthony.minessale(a)gmail.com> wrote: >> >>> Your "annoying behaviour" is the exact behavior you should be getting >>> considering what you told FS to do. >>> >>> As soon as you call uuid_bridge you are transferring both legs of the >>> call to bridge to each other. >>> This means your java app must exit so the channels can connect to each >>> other. >>> >>> remember that you hangup hook can be called when the channel is >>> transferred not only when it hangs up. >>> you have to test which is happening based on the input to your callback. >>> >>> >>> On Thu, Nov 19, 2009 at 1:46 PM, Artem Shiyanov <shiyanov(a)gmail.com>wrote: >>> >>>> Hi there! >>>> >>>> I've got annoying FS behavior: >>>> There are 2 channels executing the same Java application (application >>>> itself is an IVR). If I try to bridge them with uuid_bridged then both >>>> channels are killed. Here is a log from FS console: >>>> uuid_bridge 68587a9d-1d20-48f1-bdfc-72a2c027e1d2 >>>> 7d6c08fc-62bf-4a6c-a9ae-763d607e43de >>>> 2009-07-09 05:58:26.562783 [DEBUG] switch_ivr_bridge.c:1165 >>>> (sofia/internal/1005(a)192.168.147.130) State Change CS_EXECUTE -> >>>> CS_HIBERNATE >>>> 2009-07-09 05:58:26.562783 [DEBUG] switch_cpp.cpp:1185 hangup_hook >>>> called >>>> 2009-07-09 05:58:26.562783 [DEBUG] switch_ivr_play_say.c:1391 done >>>> playing file >>>> 2009-07-09 05:58:26.576844 [DEBUG] switch_ivr_play_say.c:1391 done >>>> playing file >>>> 2009-07-09 05:58:26.641307 [DEBUG] switch_core_session.c:933 Send signal >>>> sofia/internal/1005(a)192.168.147.130 [BREAK] >>>> 2009-07-09 05:58:26.641307 [DEBUG] switch_ivr_bridge.c:1167 >>>> (sofia/internal/1001(a)master.agent.starpoundtech.net) State Change >>>> CS_EXECUTE -> CS_HIBERNATE >>>> 2009-07-09 05:58:26.641307 [DEBUG] switch_cpp.cpp:1185 hangup_hook >>>> called >>>> API CALL [uuid_bridge(68587a9d-1d20-48f1-bdfc-72a2c027e1d2 >>>> 7d6c08fc-62bf-4a6c-a9ae-763d607e43de)] output: >>>> +OK 7d6c08fc-62bf-4a6c-a9ae-763d607e43de >>>> >>>> freeswitch(a)localhost.localdomain> 2009-07-09 05:58:26.674348 [DEBUG] >>>> switch_core_session.c:933 Send signal >>>> sofia/internal/1001(a)master.agent.starpoundtec >>>> 2009-07-09 05:58:26.714809 [DEBUG] switch_core_session.c:813 Send signal >>>> sofia/internal/1005(a)192.168.147.130 [BREAK] >>>> >>>> 2009-07-09 05:58:26.742764 [CRIT] mod_local_stream.c:234 Leaking stream >>>> handle! [switch_ivr_play_file() src/switch_ivr_play_say.c:1026] >>>> 2009-07-09 05:58:26.754791 [DEBUG] switch_core_session.c:813 Send signal >>>> sofia/internal/1001(a)master.agent.starpoundtech.net [BREAK] >>>> >>>> (FS version is 1.0.4) >>>> >>>> Any thoughts? >>>> >>>> >>>> Artem >>>> >>>> >>>> >>>> _______________________________________________ >>>> 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 >>>> http://www.freeswitch.org >>>> >>>> >>> >>> >>> -- >>> Anthony Minessale II >>> >>> FreeSWITCH http://www.freeswitch.org/ >>> ClueCon http://www.cluecon.com/ >>> Twitter: http://twitter.com/FreeSWITCH_wire >>> >>> AIM: anthm >>> MSN:anthony_minessale@hotmail.com <MSN%3Aanthony_minessale(a)hotmail.com> >>> GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com<PAYPAL%3Aanthony.minessale(a)gmail.com> >>> IRC: irc.freenode.net #freeswitch >>> >>> FreeSWITCH Developer Conference >>> sip:888@conference.freeswitch.org <sip%3A888(a)conference.freeswitch.org> >>> iax:guest@conference.freeswitch.org/888 >>> googletalk:conf+888@conference.freeswitch.org<googletalk%3Aconf%2B888(a)conference.freeswitch.org> >>> pstn:213-799-1400 >>> >>> _______________________________________________ >>> 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 >>> http://www.freeswitch.org >>> >>> >> > > _______________________________________________ > 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 > http://www.freeswitch.org > > -- Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire AIM: anthm MSN:anthony_minessale@hotmail.com <MSN%3Aanthony_minessale(a)hotmail.com> GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com<PAYPAL%3Aanthony.minessale(a)gmail.com> IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org <sip%3A888(a)conference.freeswitch.org> iax:guest@conference.freeswitch.org/888 googletalk:conf+888@conference.freeswitch.org<googletalk%3Aconf%2B888(a)conference.freeswitch.org> pstn:213-799-1400 _______________________________________________ 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 http://www.freeswitch.org

2 3

Re: [Freeswitch-dev] mod event multicast drops some events on outbound call legs ?
by Matteo 04 Dec '09

04 Dec '09

Hi, ----- "Anthony Minessale" <anthony.minessale(a)gmail.com> ha scritto: > where are you seeing it drop? > if you log them in the module is it sending them but then they are not > being recv? I've done some more detailed tests. Seems that the event is sent correctly over multicast (I can get it with wireshark), but is not received on the listener in SWITCH_MODULE_RUNTIME_FUNCTION(mod_event_multicast_runtime) function. I've attached two logs: mcast_dump.txt is the udp stream got with wireshark. All events are present here, so they're on wire. mcast_on_receiver.txt is what mod_event_multicast gets in input. Seems that the issue happens frequently, but sometimes works. Maybe some race condition in the receiving routine ? I'll try to investigate further. regards, matteo

1 0

Re: [Freeswitch-dev] mod event multicast drops some events on outbound call legs ?
by Matteo 03 Dec '09

03 Dec '09

Hi, ----- "Anthony Minessale" <anthony.minessale(a)gmail.com> ha scritto: > where are you seeing it drop? on both fs istances. > if you log them in the module is it sending them but then they are not > being recv? No, they're not even enterin the event_handler callback on mod_event_multicast > can you explain a bit more? sure. The scenario is simple: two fs on 2 different machines, on same lan. Fs compiled from trunk, standard config with mod_event_multicast compiled and enabled. Sip registrations are shared ok via multicast, so module is working. Sip domain is the same on both fs istances, to be able to register to any server and call any client. This is also working ok. Then I started looking into events sent & received via multicast. I uncommented the debug print on the mcast event receiver, and added a simple debug line on the event handler callback to print out the event type ready to be sent. Every event that's sent, is received ok. No problem here. But I noticed that the events of type PRESENCE_IN, regarding the outbound leg of a call are not sent to the multicast event handler to be propagated (no debug print at all) ONLY when the outbound channel is in early state. Previous events like EXECUTE, EXECUTE_COMPLETE, CODEC, CALL_UPDATE, etc, which happens before dialing the outboung leg are sent correctly. The "missing" events are present on the core because: * enabling presence debug, I can see them on cli * I can get them on the event socket But I don't see them entering the mcast handler to be transmitted, so they cannot be received :) As soon as the outbound leg is answered, the corresponding event is sent ok via multicast. Same when the call is closed. Only the early state does not hit the handler. regards, Matteo. the call

2 2

[Freeswitch-dev] mod event multicast drops some events on outbound call legs ?
by Matteo 03 Dec '09

03 Dec '09

Hi, I was adding "multicast presence" to FS using mod_event_multicast. Useful to have BLF over some machines :) First implementation works almost ok, but I'm facing a strange issue. Seems that mod_event_multicast drops some events, which can seen correctly over mod_event_socket. For example I can see only PRESENCE_IN events for the caller, but not for the called, when the state is early. For example: if caller is 1001 and called is 1000, over event_socket I see PRESENCE_IN updates for both during all the transaction, but over mcast I see only event initial events for 1001 (which is confirmed), and I get events for 1000 only when the call is cancelled (1000 terminated) or answere (1000 is confirmed also). Over event socket is all ok, I cann see everything. Debugging presence I can get over console all the PRESENCE_IN messages. Seems for some reason that event multicast is not able to get events for the outbound leg when is in early state? Any hints? I'll be glad to release the patch to sofia to add multicast BLF when all works :) regards, matteo.

2 1

[Freeswitch-dev] FreeSWITCH Survey: What Environment Do You Normally Use?
by Michael Collins 02 Dec '09

02 Dec '09

Hi folks, I'm doing a little survey to get an idea of what everyone prefers to use for their operating environment, like 32 vs. 64 bit, Linux vs. Windows, etc. Please log in to the main page and check out this node: http://www.freeswitch.org/node/206 Select the environment that you use the most or prefer to use. In a week or so I will send out the final tally. I'm sure the only question is who will come in second place after 64-bit CentOS/Red Hat. :) -Michael

1 0