Re: [asterisk-users] Does the Eicon/Dialogic 4BRI-8M 800-665 support the NT modus?
On 16.07.2012 04:17, Michelle Konzack wrote: Hello Armin, Am 2012-07-13 15:17:33, hacktest Du folgendes herunter: On 29.06.2012 09:00, Michelle Konzack wrote: I do not know, because I have not bougth it yet. It is only named: Eicon DIVA Server 4BRI-8M 2.0 800-665-02 version 2.0 If it is really a rev2 (2.0), then the NT mode should work. Thanks. I now have bought it. Will give it a try. Question: Must I set the whole card into NT mode or can I do this per ISDN Port? Each port can be configured individialy. I have the datasheets of the Cologne Chips here and the 2port and 4port chips can set individualy... So, I assume, it is a question of the ISDN card implementer, whether it allow to set individual ports to TE/NT mode I have the Development Kit here with some samples, but failed to build a small arme based GSM/HSPA + ISDN/Analog router... Currently I use the Vodafone Easybox 803A which support generaly what I need, except that 1) it support only one USB GSM/HSPA Stick 2) I can not access the SMS storage 3) support only Analog and ISDN Telephone and Telephone-Systems and the (most important part 4) I can not use VoIP telphones So I would like to build a GSM Router which support the above stuff and of course can run a Linux-Distribution of choice using Asterisk! To start a new Open Hardware Project I search for contributors like an Electronic Engineer and Software Developers (we need special drivers for the stuff) and I prefer to use: 1) TI OMAP 300-700 MHz ARM Microcontroller 2) =3D 128 MByte of SDRAM 3) =3D 512 MByte of NAND 4) Cologne Chip XHFC-2S4U or XHFC-4U (could be a module) 5) SiLabs Quad ProSLIC 6) Cypress 4 Port Tetra Hub 7) 8port or 12port N-Way Switch supporting PoE 8) SDHC or SDXC Card interface maybe with 9) 4 TFT Display (480x272 pixel) for status 10) Integrated 5Ah LiPoly cell (10Ah option) for Backup since the router should work some hours even if there is an electricity cut. Maybe this sounds like a Dream-Box, but is entirely technicaly possibel! And of course, I have already aquired most chips in the last 2 years... Note: I am using CadSoft Eagle under Debian to desingn PCBs and a mailinglist is already available even if unused... If you are interested in Open Hardware Development feel free to contact me, because it is time to change something. Oh, forgotten one thing: I was even thinking to use a Marvel Discovery MV78100 (800/1000/1200MHz) because it support: 1) SO-DIMM Modules 2) huge NAND Flash 3) SD-Card 4) 2 PCIe 4x ports which can be each splited into 4 ports of 1x 5) 2 SATA Drives 6) 1 Giga Ethernet Sounds very interesting! Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Does the Eicon/Dialogic 4BRI-8M 800-665 support the NT modus?
On 29.06.2012 09:00, Michelle Konzack wrote: Hello Armin Schindler, Am 2012-06-28 09:52:30, hacktest Du folgendes herunter: On 27.06.2012 18:46, Michelle Konzack wrote: Does the Eicon/Dialogic 4BRI-8M 800-665 support the NT modus? Which PCI-ID is that? I do not know, because I have not bougth it yet. It is only named: Eicon DIVA Server 4BRI-8M 2.0 800-665-02 version 2.0 If it is really a rev2 (2.0), then the NT mode should work. Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Does the Eicon/Dialogic 4BRI-8M 800-665 support the NT modus?
On 27.06.2012 18:46, Michelle Konzack wrote: Does the Eicon/Dialogic 4BRI-8M 800-665 support the NT modus? Which PCI-ID is that? Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan_capi audio weirdness
Hello Arik, On 02/14/2012 12:49 PM, Arik Raffael Funke wrote: Hi, I am trying to run asterisk on an AVM Fritz!Box Fon 7270 embedded DSL router. This works quite well after getting rid of the preinstalled phone server but I am encountering some unexpected behaviour. Background: I am using two CAPI controllers provided by the hardware - one in MSN mode for dialling out and - one in NT-mode, (DID) for the internal S0-Bus The problem is, I get no audio whatsoever until a channel is answered. Some of the symptoms of this are: - If I have an s-extension for the internal S0-Bus exten = s,1,Playtones(dial) I cannot hear the dialtone. It works however with: exten = s,1,Answer exten = s,n,Playtones(dial) - Similarly if I dial from internal to external with the extension: exten = _X.,1,Dial(CAPI/contr1/12345) I hear no progress indication. EVEN when using the r-option of the dial command. It works however with exten = _X.,1,Answer exten = _X.,n,Dial(CAPI/contr1/12345) in NT mode, the B-channel is not activated automatically. You have to signal the TE side that early-B3 data is available. Then the TE side can activate the B-channel. If the NT-side is chan_capi, use exten = _X.,n,capicommand(progress) without Answer before Dial(). Also, when using Dial() with chan_capi, you should use /b or /B option in Dial() to get early-B3 from that other side too. See README of chan_capi package for more details. Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Call holding with chan_capi
Hi,
On 02/14/2012 06:28 PM, Arik Raffael Funke wrote:
My apologies, I just realised I copied the wrong section of the debug log. So
once again, when pressing the park call button, I get the following capi
debug output:
CAPI: ApplId=0x0002 Command=0x80 SubCommand=0x82 MsgNum=0xe446 NCCI=0x1403
FACILITY_IND ID=002 #0xe446 LEN=0018
Controller/PLCI/NCCI= 0x1403
FacilitySelector= 0x3
FacilityIndicationParameter = 02 80 00
-- ISDN_INTERN#02: unhandled FACILITY_IND supplementary function 8002
FACILITY_RESP ID=002 #0xe446 LEN=0015
Controller/PLCI/NCCI= 0x1403
FacilitySelector= 0x3
FacilityResponseParameters = default
CAPI: ApplId=0x0002 Command=0x84 SubCommand=0x82 MsgNum=0xe447 NCCI=0x00011403
DISCONNECT_B3_IND ID=002 #0xe447 LEN=0015
Controller/PLCI/NCCI= 0x11403
Reason_B3 = 0x3301
NCPI= default
DISCONNECT_B3_RESP ID=002 #0xe447 LEN=0012
Controller/PLCI/NCCI= 0x11403
looks like normal hold where B-channel is released.
When you use capicommand(hold), you can specify a second parameter. This
parameter is the name of a variable which is filled with the reference ID of
the hold. capicommand(retrieve, ${HOLDID}) then can unhold the call.
See README of chan_capi package for details.
Armin
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
http://www.asterisk.org/hello
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] AVM Fritz! mISDN with Kernel 2.6.32 - Any experiences? - Email found in subject
Hi, it seems that the asterisk API here was changed again and chan_capi must be adapted to this. I will have a look. Armin On Mon, 1 Mar 2010, dle...@lstelcom.com wrote: Hi again! I have excellent success with the tiny fcpci and chan_capi, which is also working great with capi4hylafax. See net-dialup/fcpci-0.1-r1 in gentoo (should not be difficult to use this on other distros, but I have never done so). Do not confuse this with the fritzcapi! I managed to install fcpci and it seems to run fine (capiinfo output). Unfortunately i cant compile chan_capi against my Asterisk trunk r240716. Neither the trunk/head nor the 1.1.4 Version compiles. All fail with the following output: srvpbx:/usr/src/chan-capi-HEAD# make [CC] chan_capi.c - chan_capi.o In file included from chan_capi.c:32: chan_capi.h:34:26: error: asterisk/rtp.h: Datei oder Verzeichnis nicht gefunden chan_capi.c: In function âlocal_queue_frameâ: chan_capi.c:803: error: invalid operands to binary == (have âunion anonymousâ and âintâ) chan_capi.c: In function âinterface_cleanupâ: chan_capi.c:1071: warning: implicit declaration of function âast_rtp_destroyâ chan_capi.c: In function âsend_progressâ: chan_capi.c:1165: error: incompatible types in assignment chan_capi.c: In function âclear_channel_fax_loopâ: chan_capi.c:2884: error: invalid operands to binary == (have âunion anonymousâ and âintâ) chan_capi.c: In function âcapidev_handle_did_digitsâ: chan_capi.c:3548: error: incompatible types in assignment chan_capi.c: In function âcapi_queue_cause_controlâ: chan_capi.c:3564: warning: missing braces around initializer chan_capi.c:3564: warning: (near initialization for âfr.subclassâ) chan_capi.c:3569: error: incompatible types in assignment chan_capi.c:3573: error: incompatible types in assignment chan_capi.c: In function âcapidev_handle_info_indicationâ: chan_capi.c:3876: error: incompatible types in assignment chan_capi.c:3886: error: incompatible types in assignment chan_capi.c: In function âhandle_facility_indication_dtmfâ: chan_capi.c:4138: error: incompatible types in assignment chan_capi.c:4149: error: incompatible types in assignment chan_capi.c: In function âcapidev_handle_data_b3_indicationâ: chan_capi.c:4292: error: incompatible types in assignment chan_capi.c:4294: error: incompatible types in assignment chan_capi.c: In function âcapi_signal_answerâ: chan_capi.c:4316: warning: missing braces around initializer chan_capi.c:4316: warning: (near initialization for âfr.subclassâ) chan_capi.c: In function âcapidev_handle_disconnect_indicationâ: chan_capi.c:4605: warning: missing braces around initializer chan_capi.c:4605: warning: (near initialization for âfr.subclassâ) chan_capi.c:4654: error: incompatible types in assignment chan_capi.c: In function âcapidev_handle_connection_confâ: chan_capi.c:5025: warning: missing braces around initializer chan_capi.c:5025: warning: (near initialization for âfr.subclassâ) chan_capi.c: At top level: chan_capi.c:7746: warning: initialization from incompatible pointer type chan_capi.c: In function âconf_interfaceâ: chan_capi.c:8153: warning: passing argument 2 of âast_parse_allow_disallowâ from incompatible pointer type chan_capi.c:8156: warning: passing argument 2 of âast_parse_allow_disallowâ from incompatible pointer type make: *** [chan_capi.o] Fehler 1 Please excuse the messed up german output, i have yet to discover how to set a debian box to german keyboard, everything else english. I am limited to Asterisk trunk r240716 because i want to evaluate the T.38 - T.30 gateway function, which can only be patched into this revision. In my opinion there are the following alternatives in order to get the Fritz card running with Asterisk: A) Get chan-capi to compile: Unfortunately my C knowledge seems insufficent for this. I have found out that there is noch rtp.h in the asterisk source dir, only a file named rtp_engine.h. Changing the include accordingly unfortunately only fixes the very first error. I can't make any sense out of the second error (error: invalid operands to binary == (have union anonymous and int)), as, in line 803, the left-hand operand of the == is no union, let alone a anonymous union. I have already posted this at the chan-capi-users ML, but with no answer so far. B) Use a binary chan_capi.so from elsewhere?! I run Debian 5.0.2 Kernel 2.6.26-2-686. Could i take a compiled chan_capi.so from a machine with a different Asterisk 1.6 Version? C) Find another way to use fcpci eith Asterisk. Is this even possible? With mISDN? ...? Comments or any other Ideas are very appreciated. Sincerly Daniel Leese P.s.: Philipp, many thanks fo your answers so far ;) -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users --
Re: [asterisk-users] AVM Fritz! mISDN with Kernel 2.6.32 - Any experiences? - Email found in subject
Hi, chan_capi trunk should be compilable now with current asterisk trunk. Armin On Mon, 1 Mar 2010, Armin Schindler wrote: Hi, it seems that the asterisk API here was changed again and chan_capi must be adapted to this. I will have a look. Armin On Mon, 1 Mar 2010, dle...@lstelcom.com wrote: Hi again! I have excellent success with the tiny fcpci and chan_capi, which is also working great with capi4hylafax. See net-dialup/fcpci-0.1-r1 in gentoo (should not be difficult to use this on other distros, but I have never done so). Do not confuse this with the fritzcapi! I managed to install fcpci and it seems to run fine (capiinfo output). Unfortunately i cant compile chan_capi against my Asterisk trunk r240716. Neither the trunk/head nor the 1.1.4 Version compiles. All fail with the following output: srvpbx:/usr/src/chan-capi-HEAD# make [CC] chan_capi.c - chan_capi.o In file included from chan_capi.c:32: chan_capi.h:34:26: error: asterisk/rtp.h: Datei oder Verzeichnis nicht gefunden chan_capi.c: In function âlocal_queue_frameâ: chan_capi.c:803: error: invalid operands to binary == (have âunion anonymousâ and âintâ) chan_capi.c: In function âinterface_cleanupâ: chan_capi.c:1071: warning: implicit declaration of function âast_rtp_destroyâ chan_capi.c: In function âsend_progressâ: chan_capi.c:1165: error: incompatible types in assignment chan_capi.c: In function âclear_channel_fax_loopâ: chan_capi.c:2884: error: invalid operands to binary == (have âunion anonymousâ and âintâ) chan_capi.c: In function âcapidev_handle_did_digitsâ: chan_capi.c:3548: error: incompatible types in assignment chan_capi.c: In function âcapi_queue_cause_controlâ: chan_capi.c:3564: warning: missing braces around initializer chan_capi.c:3564: warning: (near initialization for âfr.subclassâ) chan_capi.c:3569: error: incompatible types in assignment chan_capi.c:3573: error: incompatible types in assignment chan_capi.c: In function âcapidev_handle_info_indicationâ: chan_capi.c:3876: error: incompatible types in assignment chan_capi.c:3886: error: incompatible types in assignment chan_capi.c: In function âhandle_facility_indication_dtmfâ: chan_capi.c:4138: error: incompatible types in assignment chan_capi.c:4149: error: incompatible types in assignment chan_capi.c: In function âcapidev_handle_data_b3_indicationâ: chan_capi.c:4292: error: incompatible types in assignment chan_capi.c:4294: error: incompatible types in assignment chan_capi.c: In function âcapi_signal_answerâ: chan_capi.c:4316: warning: missing braces around initializer chan_capi.c:4316: warning: (near initialization for âfr.subclassâ) chan_capi.c: In function âcapidev_handle_disconnect_indicationâ: chan_capi.c:4605: warning: missing braces around initializer chan_capi.c:4605: warning: (near initialization for âfr.subclassâ) chan_capi.c:4654: error: incompatible types in assignment chan_capi.c: In function âcapidev_handle_connection_confâ: chan_capi.c:5025: warning: missing braces around initializer chan_capi.c:5025: warning: (near initialization for âfr.subclassâ) chan_capi.c: At top level: chan_capi.c:7746: warning: initialization from incompatible pointer type chan_capi.c: In function âconf_interfaceâ: chan_capi.c:8153: warning: passing argument 2 of âast_parse_allow_disallowâ from incompatible pointer type chan_capi.c:8156: warning: passing argument 2 of âast_parse_allow_disallowâ from incompatible pointer type make: *** [chan_capi.o] Fehler 1 Please excuse the messed up german output, i have yet to discover how to set a debian box to german keyboard, everything else english. I am limited to Asterisk trunk r240716 because i want to evaluate the T.38 - T.30 gateway function, which can only be patched into this revision. In my opinion there are the following alternatives in order to get the Fritz card running with Asterisk: A) Get chan-capi to compile: Unfortunately my C knowledge seems insufficent for this. I have found out that there is noch rtp.h in the asterisk source dir, only a file named rtp_engine.h. Changing the include accordingly unfortunately only fixes the very first error. I can't make any sense out of the second error (error: invalid operands to binary == (have union anonymous and int)), as, in line 803, the left-hand operand of the == is no union, let alone a anonymous union. I have already posted this at the chan-capi-users ML, but with no answer so far. B) Use a binary chan_capi.so from elsewhere?! I run Debian 5.0.2 Kernel 2.6.26-2-686. Could i take a compiled chan_capi.so from a machine with a different Asterisk 1.6 Version? C) Find another way to use fcpci eith Asterisk. Is this even possible? With mISDN? ...? Comments or any other Ideas are very appreciated. Sincerly Daniel Leese P.s.: Philipp, many thanks fo your answers so far ;) -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list
Re: [asterisk-users] SIP RTP ports not released when channel is hung up
On Tue, 16 Feb 2010, Armin Schindler wrote: On Tue, 16 Feb 2010, Marcus Hunger wrote: Hi, did you see this one: https://issues.asterisk.org/view.php?id=16774 ? It looks related to your issue. Oh thanks, I missed that one. It really looks related. I have added a note. Now I know how to reproduce the problem. I added this as note to 16774 as well: Start SIP client to register at asterisk, then disconnect that SIP phone from network. In the time the registration is still active in asterisk, call this phone. Asterisk will send INVITEs (of course with no answer), then hangup after about 30 seconds. The asterisk channels are released, but the sip channel for that Init: INVITE is not released. For now, I can confirm this with 1.4.28 only as I have not tested other versions yet. Armin Best regards, Marcus On Fri, Feb 12, 2010 at 12:04 PM, Armin Schindler ar...@melware.de wrote: On Fri, 12 Feb 2010, Armin Schindler wrote: I had a look at netstat -nuap and it shows that a lot of ports are still assigned, even if there is no channel in use. But sip show channels show a lot of (unused) entries with no codec/Format and Last Message like INVITE, REGISTER, OPTIONS. REGISTER and OPTIONS allocate no RTP ports, so those are not a problem. If you have a SIP channel that has a last message being INVITE and still say you have no calls, you have a problem right there. I just see these entries with sip show channels, but cannot tell if e.g. the REGISTER listed channels have RTP ports allocated. Who can I find out which SIP channel allocated which port? Or which SIP channel belongs to the ports I see with 'netstat -nuap'? I just made a test to confirm: After a restart of asterisk (to have a clean state with no sip channels activ and no RTP port allocated), I can confirm that: - REGISTER and OPTION listed sip channels don't use RTP ports - after some calls (e.g. SIP to SIP) the RTP ports are freed immediately (looks like this is the case on hangup before answer). - after some other calls, the RTP ports are freed after about 20-30 seconds after hangup. So basically all is correct. I do have a sip channels like 172.21.4.114 666 0430c3a638e 00102/0 0x0 (nothing) No Init: INVITE in 'sip show channels' and they don't go away for a long time. Shouldn't there be a timeout to destroy such a channel even if somehow the phone was 'disconnected' in during a call? If the channels exists even after 32 seconds after BYE, and BYE was signaled correctly, I would file a bug report. It really looks like that there is a case where the sip channel is not destroyed and that is the cause of the problem. I will try to reproduce this. Any ideas? Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- Dipl.-Inf. (FH) Marcus Hunger - hun...@sipgate.de Telefon: +49 (0)211-63 55 55-61 Telefax: +49 (0)211-63 55 55-22 sipgate GmbH - Gladbacher Str. 74 - 40219 Düsseldorf HRB Düsseldorf 39841 - Geschäftsführer: Thilo Salmon, Tim Mois Steuernummer: 106 / 5724 / 7147, Umsatzsteuer-ID: DE219349391 www.sipgate.de - www.sipgate.at - www.sipgate.co.uk -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] SIP RTP ports not released when channel is hung up
On Thu, 18 Feb 2010, Karsten Wemheuer wrote: Am Donnerstag, den 18.02.2010, 10:49 +0100 schrieb Armin Schindler: On Tue, 16 Feb 2010, Armin Schindler wrote: On Tue, 16 Feb 2010, Marcus Hunger wrote: Hi, did you see this one: https://issues.asterisk.org/view.php?id=16774 ? It looks related to your issue. Oh thanks, I missed that one. It really looks related. I have added a note. Now I know how to reproduce the problem. I added this as note to 16774 as well: Start SIP client to register at asterisk, then disconnect that SIP phone from network. In the time the registration is still active in asterisk, call this phone. Asterisk will send INVITEs (of course with no answer), then hangup after about 30 seconds. The asterisk channels are released, but the sip channel for that Init: INVITE is not released. For now, I can confirm this with 1.4.28 only as I have not tested other versions yet. With version 1.4.29 I can't reproduce it the way You described it. If the caller hangs up before * times out the INVITE, the ressources are freed (SIP-channel and RTP-Ports). If * times out first, the ressources are freed some time later ( 1 minute). Yes, I can confirm that. I now have updated the production system to 1.4.29 and the issue seems to be solved. I cannot reproduce it anymore. Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] SIP RTP ports not released when channel is hung up
On Tue, 16 Feb 2010, Marcus Hunger wrote: Hi, did you see this one: https://issues.asterisk.org/view.php?id=16774 ? It looks related to your issue. Oh thanks, I missed that one. It really looks related. I have added a note. Thanks, Armin Best regards, Marcus On Fri, Feb 12, 2010 at 12:04 PM, Armin Schindler ar...@melware.de wrote: On Fri, 12 Feb 2010, Armin Schindler wrote: I had a look at netstat -nuap and it shows that a lot of ports are still assigned, even if there is no channel in use. But sip show channels show a lot of (unused) entries with no codec/Format and Last Message like INVITE, REGISTER, OPTIONS. REGISTER and OPTIONS allocate no RTP ports, so those are not a problem. If you have a SIP channel that has a last message being INVITE and still say you have no calls, you have a problem right there. I just see these entries with sip show channels, but cannot tell if e.g. the REGISTER listed channels have RTP ports allocated. Who can I find out which SIP channel allocated which port? Or which SIP channel belongs to the ports I see with 'netstat -nuap'? I just made a test to confirm: After a restart of asterisk (to have a clean state with no sip channels activ and no RTP port allocated), I can confirm that: - REGISTER and OPTION listed sip channels don't use RTP ports - after some calls (e.g. SIP to SIP) the RTP ports are freed immediately (looks like this is the case on hangup before answer). - after some other calls, the RTP ports are freed after about 20-30 seconds after hangup. So basically all is correct. I do have a sip channels like 172.21.4.114 666 0430c3a638e 00102/0 0x0 (nothing) No Init: INVITE in 'sip show channels' and they don't go away for a long time. Shouldn't there be a timeout to destroy such a channel even if somehow the phone was 'disconnected' in during a call? If the channels exists even after 32 seconds after BYE, and BYE was signaled correctly, I would file a bug report. It really looks like that there is a case where the sip channel is not destroyed and that is the cause of the problem. I will try to reproduce this. Any ideas? Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users -- Dipl.-Inf. (FH) Marcus Hunger - hun...@sipgate.de Telefon: +49 (0)211-63 55 55-61 Telefax: +49 (0)211-63 55 55-22 sipgate GmbH - Gladbacher Str. 74 - 40219 Düsseldorf HRB Düsseldorf 39841 - Geschäftsführer: Thilo Salmon, Tim Mois Steuernummer: 106 / 5724 / 7147, Umsatzsteuer-ID: DE219349391 www.sipgate.de - www.sipgate.at - www.sipgate.co.uk -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] SIP RTP ports not released when channel is hung up
using Asterisk 1.4.28, I encountered a problem with SIP RTP port allocation. I found some entries in mailinglist and bugtracker regarding this issue, but only old ones. My rtp.conf has [general] rtpstart=3 rtpend=30100 so 100 ports available. I know that up to 4 ports per channel can be used and so up to 25 channels are possible. 4 ports only if you use audio and video. We use two ports per RTP stream - and send on two ports, but this is for incoming media. So 100 ports is enough for 50 audio calls. Even if it isn't a video call, I think as soon as videosupport is activated, the additional 2 ports are allocated. But even earlier I often get the error about No RTP ports remaining. I had a look at netstat -nuap and it shows that a lot of ports are still assigned, even if there is no channel in use. But sip show channels show a lot of (unused) entries with no codec/Format and Last Message like INVITE, REGISTER, OPTIONS. REGISTER and OPTIONS allocate no RTP ports, so those are not a problem. If you have a SIP channel that has a last message being INVITE and still say you have no calls, you have a problem right there. I just see these entries with sip show channels, but cannot tell if e.g. the REGISTER listed channels have RTP ports allocated. Who can I find out which SIP channel allocated which port? Or which SIP channel belongs to the ports I see with 'netstat -nuap'? I do have a sip channels like 172.21.4.1146660430c3a638e 00102/0 0x0 (nothing)No Init: INVITE in 'sip show channels' and they don't go away for a long time. Shouldn't there be a timeout to destroy such a channel even if somehow the phone was 'disconnected' in during a call? If the channels exists even after 32 seconds after BYE, and BYE was signaled correctly, I would file a bug report. Yes, the RTP ports should be closed at least at that point, when we destroy the SIP channel. Anything else is a bug. I am not really sure about when they're closed, but I'm trying to understand that in my RTCP adventures since I want to change it. Before filing a bug, I would like to be sure that I have checked all possibilities here. To me it looks like that some special event leaves a sip channel activ and not be destroyed. So when Asterisk runs for a longer time, more and more channels like this occur. Ayn idea how to check this? Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] SIP RTP ports not released when channel is hung up
On Fri, 12 Feb 2010, Armin Schindler wrote: I had a look at netstat -nuap and it shows that a lot of ports are still assigned, even if there is no channel in use. But sip show channels show a lot of (unused) entries with no codec/Format and Last Message like INVITE, REGISTER, OPTIONS. REGISTER and OPTIONS allocate no RTP ports, so those are not a problem. If you have a SIP channel that has a last message being INVITE and still say you have no calls, you have a problem right there. I just see these entries with sip show channels, but cannot tell if e.g. the REGISTER listed channels have RTP ports allocated. Who can I find out which SIP channel allocated which port? Or which SIP channel belongs to the ports I see with 'netstat -nuap'? I just made a test to confirm: After a restart of asterisk (to have a clean state with no sip channels activ and no RTP port allocated), I can confirm that: - REGISTER and OPTION listed sip channels don't use RTP ports - after some calls (e.g. SIP to SIP) the RTP ports are freed immediately (looks like this is the case on hangup before answer). - after some other calls, the RTP ports are freed after about 20-30 seconds after hangup. So basically all is correct. I do have a sip channels like 172.21.4.1146660430c3a638e 00102/0 0x0 (nothing)No Init: INVITE in 'sip show channels' and they don't go away for a long time. Shouldn't there be a timeout to destroy such a channel even if somehow the phone was 'disconnected' in during a call? If the channels exists even after 32 seconds after BYE, and BYE was signaled correctly, I would file a bug report. It really looks like that there is a case where the sip channel is not destroyed and that is the cause of the problem. I will try to reproduce this. Any ideas? Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[asterisk-users] SIP RTP ports not released when channel is hung up
Hello, using Asterisk 1.4.28, I encountered a problem with SIP RTP port allocation. I found some entries in mailinglist and bugtracker regarding this issue, but only old ones. My rtp.conf has [general] rtpstart=3 rtpend=30100 so 100 ports available. I know that up to 4 ports per channel can be used and so up to 25 channels are possible. But even earlier I often get the error about No RTP ports remaining. I had a look at netstat -nuap and it shows that a lot of ports are still assigned, even if there is no channel in use. But sip show channels show a lot of (unused) entries with no codec/Format and Last Message like INVITE, REGISTER, OPTIONS. Why aren't RTP ports released when not in use? Or is there a possibility to configure this behaviour? Thanks, Armin -- _ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] G.722 problems with IAX
On Mon, 14 Sep 2009, Stephen Davies wrote: 2009/9/9 Armin Schindler ar...@melware.de No, I didn't miss that. See my text. I mentioned this because I think this might be the reason of the problem and the incorrect handling in jitterbuffer, if it is the jitterbuffer. It is just a guess, since everything else seems to work good. The question is why does G.722 via IAX has problems. Is anyone using it and can say it works in his setup? I'm not sure if Steve Kann is still around the project, but if not, I'm familiar with chan_iax2.c and mostly familiar with the iax2 jitter buffer so I might have a go at fixing the problem. Will you open a bug on the bugs.digium.com bug tracker. I did do a test from a SNOM820 (yum) via an IAX trunk with jitter buffer and got the same nasty jerky audio. This is a recent checkout of branch-1.4. I have opened the bug 0015901 and added the link to the patch I use to have g.722 support with asterisk-1.4.25 (gain problem on transcoding is fixed here). Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- AstriCon 2009 - October 13 - 15 Phoenix, Arizona Register Now: http://www.astricon.net asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] G.722 problems with IAX
On Fri, 4 Sep 2009, Tim Panton wrote: On 4 Sep 2009, at 07:53, Armin Schindler wrote: On Thu, 3 Sep 2009, Tilghman Lesher wrote: On Thursday 03 September 2009 02:47:05 Armin Schindler wrote: Hello, I try to move our asterisk installation (3 Asterisk servers in different offices connected using IAX and a lot of SIP phones, as well as ISDN connections using CAPI) to use G.722 instead of G.711. Asterisk 1.4.25.1 is used with the G.722 patch (the fixed one, which solves the gain problem). So SIP-to-SIP and to ISDN there is no problem. G.722 itself works and transconding to G.711 for ISDN also works good. But when I make a connection through IAX to another asterisk (having allow=g722 to really use G.722 in IAX) the voice is 'broken'. I also work on G.722 for twinklephone and encountered a special thing about G.722: It has a sample rate of 16000, but it announced as 8000 in SDP. Since I have similar problem with my G.722-twinkle implementation, it looks like the RTP and/or jitterbuffer code has a problem with that. Did I miss something here or is this really a bug? You missed that the IETF has a typo in the specification, stating that G.722 is to be stated as 8000, even though it's 16000. This will remain, due to backwards compatibility concerns. Please see RFC 3551, section 4.5.2. http://www.apps.ietf.org/rfc/rfc3551.html#sec-4.5.2 No, I didn't miss that. See my text. I mentioned this because I think this might be the reason of the problem and the incorrect handling in jitterbuffer, if it is the jitterbuffer. It is just a guess, since everything else seems to work good. The question is why does G.722 via IAX has problems. Is anyone using it and can say it works in his setup? Armin I've got g722 running through 1.4.22.2 with the patch set that targets 1.4.7 Calls from our java iax softphone come in as IAX2 in g722 and leave via SIP to a g722 conference service. seems to work ok. No transcoding, recording etc, and the jitterbuffer is _off_ since it's a VoIP to VoIP call. Yes, when is disable jitterbuffer (I had even forcejitterbuffer=yes in iax.conf and jbforce=yes in sip.conf), it works here too. I don't know how the jitterbuffer is doing it, but could it be possible that the RTP info of g722 (stated 8000 but it actually 16000) is confusing the jitterbuffer? Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- AstriCon 2009 - October 13 - 15 Phoenix, Arizona Register Now: http://www.astricon.net asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] G.722 problems with IAX
On Thu, 3 Sep 2009, Tilghman Lesher wrote: On Thursday 03 September 2009 02:47:05 Armin Schindler wrote: Hello, I try to move our asterisk installation (3 Asterisk servers in different offices connected using IAX and a lot of SIP phones, as well as ISDN connections using CAPI) to use G.722 instead of G.711. Asterisk 1.4.25.1 is used with the G.722 patch (the fixed one, which solves the gain problem). So SIP-to-SIP and to ISDN there is no problem. G.722 itself works and transconding to G.711 for ISDN also works good. But when I make a connection through IAX to another asterisk (having allow=g722 to really use G.722 in IAX) the voice is 'broken'. I also work on G.722 for twinklephone and encountered a special thing about G.722: It has a sample rate of 16000, but it announced as 8000 in SDP. Since I have similar problem with my G.722-twinkle implementation, it looks like the RTP and/or jitterbuffer code has a problem with that. Did I miss something here or is this really a bug? You missed that the IETF has a typo in the specification, stating that G.722 is to be stated as 8000, even though it's 16000. This will remain, due to backwards compatibility concerns. Please see RFC 3551, section 4.5.2. http://www.apps.ietf.org/rfc/rfc3551.html#sec-4.5.2 No, I didn't miss that. See my text. I mentioned this because I think this might be the reason of the problem and the incorrect handling in jitterbuffer, if it is the jitterbuffer. It is just a guess, since everything else seems to work good. The question is why does G.722 via IAX has problems. Is anyone using it and can say it works in his setup? Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- AstriCon 2009 - October 13 - 15 Phoenix, Arizona Register Now: http://www.astricon.net asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[asterisk-users] G.722 problems with IAX
Hello, I try to move our asterisk installation (3 Asterisk servers in different offices connected using IAX and a lot of SIP phones, as well as ISDN connections using CAPI) to use G.722 instead of G.711. Asterisk 1.4.25.1 is used with the G.722 patch (the fixed one, which solves the gain problem). So SIP-to-SIP and to ISDN there is no problem. G.722 itself works and transconding to G.711 for ISDN also works good. But when I make a connection through IAX to another asterisk (having allow=g722 to really use G.722 in IAX) the voice is 'broken'. I also work on G.722 for twinklephone and encountered a special thing about G.722: It has a sample rate of 16000, but it announced as 8000 in SDP. Since I have similar problem with my G.722-twinkle implementation, it looks like the RTP and/or jitterbuffer code has a problem with that. Did I miss something here or is this really a bug? Thanks, Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- AstriCon 2009 - October 13 - 15 Phoenix, Arizona Register Now: http://www.astricon.net asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Eicon Diva 2.01 PCI Passive BRI ISDN card
On Mon, 6 Apr 2009, Tzafrir Cohen wrote:
On Sun, Apr 05, 2009 at 11:35:18PM +0200, Puskás Zsolt wrote:
On Sunday 05 April 2009 21.28.48 Gergo Csibra wrote:
Saturday, April 4, 2009, 3:13:12 PM, Puskás wrote:
Got it working with Asterisk 1.2 installed on the same PC as Asterisk 1.4
[ in different directorys and username of course :) ] . Using isdn4linux
kernel module and Dial(Modem/ttyI0/1234567:${EXTEN}) command.
Használj MISDN-t, és ne toppostolj.
Use MISDN, and do not toppost!
I won't toppost again but you should read my first e-mail again. I got
a passive diva isdn card which are not supported:
pc:~# mISDN scan
0 mISDN compatible device(s) found:
chan_modem is deprecated. Any chance this works with chan_capi?
Yes, but you need to use an older Version of Dialogic/Eicon driver package
where support of these old cards was still part of. The Binary drivers Diva
PCI cards as well as the Diva PRO supported full CAPI.
Armin
___
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] D channel signalling, while B channels busy?
On Wed, 16 Jul 2008, Stefan-Michael Guenther wrote: Hi, I'm running a number of asterisk installations with EICON/DIALOGIC and AVM cards, together with different versions of chan_capi. All installations share a strange phenomenon: Although all B channels are busy, another incoming call to a free phone lets this phone ring. When I pickup up the phone, the caller suddenly hears the busy tone and I hear nothing. Could it be, that the D channel is forwarding a call, although it could never be accepted? Is there a way to change this behaviour because to me it doesn't make sense and it is rather annoying experience for the caller. This is a feature of your line: call-waiting. chan-capi signals this call with a special setting in variable BCHANNELINFO (see README) and you can decide what to do (reject, deflect, ...) Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- AstriCon 2008 - September 22 - 25 Phoenix, Arizona Register Now: http://www.astricon.net asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] e164.org
On Sat, 29 Mar 2008, Grey Man wrote: Does anyone know if the e164.org ENUM service is still active? If anyone who has anything to do with the e164.org ENUM site monitors this list could you check your signup page as the Captcha's (the test to see if you are human) fails for both the text and audio tests every time. I'd post a message on the e164.org forums but the signup page there has the test missing altogether. I don't really know the 'official' status, but I use it and it does work without problems. Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Can't dial out from SIP to CAPI
On Tue, 5 Feb 2008, Sebastian Pape wrote:
Hi,
I've been trying to configure my extensions.conf and sip.conf for two days
now and I'm pretty sure it's just a small typo or anything I can't find by
myself.
My setup:
- Asterisk connected via Fritz! PCI Card to a HiPath 3500 (2 channels)
- Callcentric.com SIP channel to dial out to foreign countries
- Cisco 7912 attached to asterisk using SIP (in another city)
When I dial extension 85 my Cisco phone is ringing and I can talk and
everything works fine. But when I try to dial an extension from the dialplan
I never get a connection.
I've posted my capi.conf, extensions.conf and sip.conf here:
http://pastebin.com/f19940490
Your Dial() line for CAPI:
Dial(isdn/g1/@13:${EXTEN},30,r)
is not correct.
I assume you use a newer chan_capi, then it should look like:
Dial(CAPI/g1/13:${EXTEN}/b,30)
Armin
___
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] ISDN channels not properly released after call
On Sat, 12 Jan 2008, Antonios Tsakiridis wrote: Hello everyone, I'm using very simple setup to make and receive external ISDN calls through a softphone (x-lite version 3.0 - Win32) via an asterisk box. Hardware setup: - Dialogic Diva BRI (lspci yields: Network controller: Eicon Technology Corporation DIVA Server BRI-2M/-2F (rev 01)) - ISDN BRI line Software setup: - Redhat 9 - asterisk-1.4.16.2 - chan_capi-1.0.2 ... atlas*CLI capi show channels CAPI B-channel information: Line-Name NTmode state i/o bproto isdnstate ton number ISDN1#02 nodiscP O trans *P 0x00 '1000'-'210498' ISDN1#01 nodiscP O trans * 0x00 '1000'-'210498' From there on I cannot make any other calls because there are no available channels: What version of the divas driver do you use? And what is the configuration of the BRI card (protocol)? It looks like card/isdn is not responding to the hangup command, so can you please provide a log with set verbose 5 capi debug to see what is going on? Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[asterisk-users] OT: reverse DNS error for lists.digium.com
Hello, sorry for beeing off-topic here. But can anyone confirm that there is a problem reverse resolving lists.digium.com (216.207.245.17) ? Because of this problem, my mail server will not accept mails from lists.digium.com (it is configured to accept valid DNS only). Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] OT: reverse DNS error for lists.digium.com
On Mon, 14 Jan 2008, Gaëtan Minet wrote: Hello Same problem here. That could explain why I'm asked for a subscription confirmation every day due to excessive bounces only for this list. That is exactly the problem I have. I did a whitelisting for the moment too. It seems that the lower addresses in 216.207.245. are not listed. Armin ;; Got SERVFAIL reply from 195.238.2.21, trying next server ;; Got SERVFAIL reply from 195.238.2.21, trying next server Server: 195.238.2.22 Address:195.238.2.22#53 ** server can't find 17.245.207.216.in-addr.arpa: SERVFAIL dig 245.207.216.in-addr.arpa ns ; DiG 9.4.1-P1 245.207.216.in-addr.arpa ns ;; global options: printcmd ;; Got answer: ;; -HEADER- opcode: QUERY, status: NOERROR, id: 29194 ;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 2 ;; QUESTION SECTION: ;245.207.216.in-addr.arpa. IN NS ;; ANSWER SECTION: 245.207.216.in-addr.arpa. 86400 IN NS ns1.apid.com. 245.207.216.in-addr.arpa. 86400 IN NS ns2.apid.com. ;; ADDITIONAL SECTION: ns2.apid.com. 3019IN A 63.238.52.2 ns1.apid.com. 3019IN A 63.238.52.1 ;; Query time: 142 msec ;; SERVER: 195.238.2.21#53(195.238.2.21) ;; WHEN: Mon Jan 14 17:22:26 2008 ;; MSG SIZE rcvd: 118 nslookup 216.207.245.1 ns1.apid.com Server: ns1.apid.com Address:63.238.52.1#53 ** server can't find 1.245.207.216.in-addr.arpa: SERVFAIL On 14/01/2008, at 15:58, Armin Schindler wrote: Hello, sorry for beeing off-topic here. But can anyone confirm that there is a problem reverse resolving lists.digium.com (216.207.245.17) ? Because of this problem, my mail server will not accept mails from lists.digium.com (it is configured to accept valid DNS only). Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] OT: reverse DNS error for lists.digium.com
On Mon, 14 Jan 2008, Kevin P. Fleming wrote: Armin Schindler wrote: sorry for beeing off-topic here. But can anyone confirm that there is a problem reverse resolving lists.digium.com (216.207.245.17) ? Our IT department reports that this has been corrected. Sorry, but I cannot confirm that. Armin ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Distorted audio over Eicon Diva Server BRI
On Tue, 8 Jan 2008, CSB wrote: We are experiencing slightly distorted audio with playing of recordings on our Asterisk server when the call comes in over our Eicon Diva Server BRI card. An example is an incoming call to IVR and playing some of the standard Asterisk voice prompts. Note that there is no audio problem with internal access to the same recording. Neither is there a problem with calls not involving the playing of recordings. The problem occurs consistently and is not related to system load. According to Eicon support: Asterisk is sending it's data packets (160 bytes each, i.e. 20ms) in too large intervals. This causes the transmitter of the Diva Server card to underrun and thus to fill with idle samples in regular intervals. It's almost between any two packets where we have to insert samples. ... I wonder if anyone could provide any advice on how to continue troubleshooting this issue? I never heard of that problem before. Which versions of asterisk and chan_capi (I assume you use chan_capi) do you use? If possible, can you provide a trace with set verbose 9 capi debug to me directly (not on the list, it is very big). Also, a full ditrace would help too. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Strange ISDN-problem with incoming calls out of the same city
On Thu, 13 Dec 2007, Stefan Guenther wrote: Hi, Armin wrote: How does your dialplan look like? If you have e.g. exten = _.,1, in the context for capi incoming calls, then asterisk (chan-capi) accept these calls even if not all numbers are dialed (transmitted) yet. we found the reason for our problem, but not yet the solution, maybe you have an idea. When we call the the asterisk from a mobile phone, the dialed number is transfered in one block, so asterisk sees the whole number and everything is fine. When you use a normal phone, the digits you have dialed are often not transfered in one block, but one after the other. If this transfer takes too long, the isdn card/the driver/the capi doesn't get the whole number and only transfers those digits that have already been receivec. Is there a chance to tell the isdn card/the driver/the capi to wait longer? BTW: Which component (capi/driver/card) sets this timeout? Again, same question. How does your dialplan look like? If you have a rule _. (which means 4 digits OR MORE), then there is a match even if other digits follow. Make sure your dialplan will accept the call only if all digits are dialed. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Strange ISDN-problem with incoming calls out of the same city
On Thu, 13 Dec 2007, Stefan Guenther wrote:
Hi,
Armin wrote:
Again, same question. How does your dialplan look like?
If you have a rule _. (which means 4 digits OR MORE), then there
is a match even if other digits follow. Make sure your dialplan
will accept the call only if all digits are dialed.
here is the relevant part:
; Abschnitt fuer das Analogfax
exten = 940325,1,ANSWER()
exten = 940325,2,WAIT(1)
exten = 940325,3,DIAL(CAPI/g2/${EXTEN:4},20,tr)
exten = 940325,4,HANGUP()
; Alle weiteren Anschluesse (SIP)
exten = _9403XX,1,ANSWER()
exten = _9403XX,2,SET(CHANNEL(language)=de))
exten = _9403XX,3,DIAL(${SIP${EXTEN:4}},30,tr)
exten = _9403XX,4,HANGUP()
That's how we started. We then had incoming calls, e.g. 94033 according
to the cdr logfile. But most of the calls come in with the correct
number e.g. 940331.
We have one caller where the transfered number is always to short.
Everytime he dials 930331, the number comes in as 94033. So I inserted
exten = 940331,1,ANSWER()
exten = 940331,2,WAIT(1)
exten = 940331,3,DIAL(CAPI/g2/${EXTEN:4},20,tr)
exten = 940331,4,HANGUP()
at the beginning of the file, but still the call came in as 94033.
That looks corrct so far, numbers with just 94033 should not be accepted,
because of no match.
What type of ISDN line do you have?
And how is chan-capi (capi.conf) configured?
BTW, why do you Answer() all calls directly?
Armin
___
--Bandwidth and Colocation Provided by http://www.api-digital.com--
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Strange ISDN-problem with incoming calls out of the same city
On Thu, 13 Dec 2007, Stefan Guenther wrote: Hi, Armin wrote: That looks corrct so far, numbers with just 94033 should not be accepted, because of no match. I then added a context to match those, too. Otherwise I would loose these calls. What type of ISDN line do you have? 4 BRI (Anlagenanschluß) connected to an EICON 4 BRI8M Okay, but then the setting in capi.conf isdnmode=MSN is wrong. For 'Anlagenanschluss' you need isdnmode=did And how is chan-capi (capi.conf) configured? [General] nationalprefix=0 internationalprefix=00 rxgain=0.8 txgain=0.8 language=deA These options are no general options, these are for the port sections only: immediate=yes ;immediate=yes faxdetect=off [ISDN1] incomingmsn=* context=from-extern isdnmode=MSN controller=1 group=1 callgroup=1 accountcode=ISDN1 echocancel=yes echosquelch=1 Both of these do not make sense. Either your hardware does not provide echo-cancel and you use echosquelch (which is something I cannot recommend) or you have a DIVA with hardware echo-cancel and you set echocancel=yes. echotail=64 devices=2 [cut] ISDN3-5 like ISDN1 ISDN2 is in NT mode. BTW, why do you Answer() all calls directly? Hm, I was used to put an ANSWER() in front of every PLAYBACK(), BACKGROUND(), VOICEMAILMAIN() and so on. And after I while I asked myself whether it wouldn't be a good idea to put ANSWER() in front of DIAL(), too, just to be sure that asterisk has control of the call. No good idea? Could it cause an trouble? The wiki doesn't mention any disadvantages. If you want to play something to the caller (without having reversed early-B3 possibility), you need to ANSWER(). But if the caller just should hear the ringing tone, let Dial() handle this. E.g., if the extension is busy, the caller must pay for the connection because of the ANSWER(). Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Strange ISDN-problem with incoming calls out of the same city
On Thu, 13 Dec 2007, Steve Totaro wrote:
Tzafrir Cohen wrote:
On Thu, Dec 13, 2007 at 10:26:59AM -0500, Steve Totaro wrote:
Stefan Guenther wrote:
exten = 940331,3,DIAL(CAPI/g2/${EXTEN:4},20,tr)
How do your zap conf files look?
chan_capi is used in this thread.
I have never used chan_capi, just trying to help. Are there settings
such as overlapdialing such as chan_zap?
As far as I know overlapdialing is for outgoing calls on ZAP only, right?
chan-capi does not have this option. If you want to give additional digits
in/after Dial(), you just use the /o option in Dial() command.
Armin
___
--Bandwidth and Colocation Provided by http://www.api-digital.com--
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Strange ISDN-problem with incoming calls out of the same city
On Thu, 13 Dec 2007, Stefan Guenther wrote: Hi, Armin wrote: Okay, but then the setting in capi.conf isdnmode=MSN is wrong. For 'Anlagenanschluss' you need isdnmode=did Okay , I changed that. These options are no general options, these are for the port sections only: immediate=yes The example capi.conf says about this parameter: ;immediate=yes ;DID: immediate start of pbx with extension 's' if no ; digits were received on incoming call (no destination ; number yet) ;MSN: start pbx on CONNECT_IND and don't wait for ; SETUP/SENDING-COMPLETE. info like REDIRECTINGNUMBER ; may be lost, but this is necessary for ; drivers/pbx/telco which does not send SETUP or ; SENDING-COMPLETE. Since I still face the problem, that asterisk doesn't get all digits of a telephone number, do you think that switching from MSN to DID will have any influence on this? Because with MSN the pbx wont' find for a SEDNING-COMPLETE? Sounds to me as if the pbx would react on incomplete numbers, too, which is exactly me problem! If you have a DID line (Anlagenanschluss), you need to set isdnmode=did. Otherwise chan-capi will not wait for additional digits. In DID mode, chan-capi will collect all digits until a match is found in the dialplan. If a match would be possible if further digits are received, chan-capi will wait. If a match is not possible any more, the call is now ignored by chan-capi (maybe another CAPI application is interested). Thanks for the hint on ANSWER(). Although it might keep people from calling me day night, because they would have to pay anyway, even when the phone is busy ;-)) If they know that they have to pay ;-) Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Strange ISDN-problem with incoming calls out of the same city
On Fri, 7 Dec 2007, Stefan Guenther wrote: Hi, Does this number (you are dialing) has been ported from a different Telco? When you dial from the other city and you get service not available you may be dialing from a different Telco that either has no route aggreement for the dialed network, or the number portability database (of Out of city Operator) is not up to date. before we switched from the old pbx to the asterisk server, these people had no problems calling our client. With some more debugging we saw what happens with these specific calls. For some reason local calls and calls from a few other cities cause trouble, because asterisk doesn't get the whole number that has been dialed. If e.g. someone from the same town dials 123456, asterisk only gets 12345 or 1234. This extension doesn't exist in the dialplan and so the call fails. And this is not a single failure, it happens every time. The telco has checked the lines and they are okay, so it might be the ISDN card (EICON) or the driver. I have made a trace log from one of these failed calls and will forward it to EICON. Meanwhile we catch all these calls with the i extension. How does your dialplan look like? If you have e.g. exten = _.,1, in the context for capi incoming calls, then asterisk (chan-capi) accept these calls even if not all numbers are dialed (transmitted) yet. Anyway, you talk about external calls, but you set ntmode=yes which does not make sense. Also, you should set isdnmode= to whatever isdn mode you have on your line. Please have a look at the example capi.conf of chan-capi package. Some of your general settings are possible in the interface sections only. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Little OT: Compilation of EICON driver fails with capi errors
On Sat, 27 Oct 2007, Stefan Guenther wrote: Hello, I want to use a 4 port ISDN card from EICON/DIALOGIC in our asterisk server. The system is a Ubuntu 7.10 with kernel 2.6.23.1. The compilation of the kernel finishes without any problems. I have downloaded and installed the deb-source package that EICON/DIALOGIC offers. Th installation script crashes with the following error messages: ... drivers/isdn/capi/kcapi.c:1014:47: error: macro INIT_WORK passed 3 arguments, but takes just 2 make[2]: *** [drivers/isdn/capi/kcapi.o] Error 1 make[1]: *** [drivers/isdn/capi] Error 2 make: *** [_module_drivers/isdn] Error 2 #+ LOG INFO: pwd:/usr/lib/eicon/divas/src #! LOG ABORT EXECUTION DUE TO ERROR : Failed to call 'make modules' #! LOG ERROR INFO: make modules Has anyone on the list experienced similar problems with an EICON card and has found a solution? I also tested the installation with kernel 2.6.22-14 which is the one that comes with Ubuntu - same problem. I already contacted EICON support but they haven't answered yet. It looks like the kcapi module that's coming with the Eicon package is incompatible with the new kernel version. If you don't have the need for using the Eicon package, you might want to try the Melware V3 driver which uses the in-kernel capi and will not patch the kernel on your system. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Little OT: Compilation of EICON driver fails with capi errors
On Sat, 27 Oct 2007, Patrick wrote: On Sat, 2007-10-27 at 19:49 +0200, Stefan Guenther wrote: Hello, I want to use a 4 port ISDN card from EICON/DIALOGIC in our asterisk server. The system is a Ubuntu 7.10 with kernel 2.6.23.1. The compilation of the kernel finishes without any problems. I have downloaded and installed the deb-source package that EICON/DIALOGIC offers. Th installation script crashes with the following error messages: [snip] I have a couple of single BRI and one quad BRI Eicon Diva Server cards. On Fedora 6 and 7 and CentOS 4.x and 5 there is no need to install anything from Eicon. The kernel already includes the modules for these Eicon Diva Server cards. Here is how I load the modules manually: ... Yes, if you don't any of the new features, you can go with the V2 driver, which is part of the kernel. But newer cards need the new driver. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Eicon Diva and incoming Fax
Hello VoipCrazy !? On Mon, 15 Oct 2007, voip crazy wrote: Hello all, I am trying to set up asterisk and hylafax to send and receibe fax. The machine is connected to the PSTN is an Eicon Diva BRI (1 BRI port). My problem is that , when I send a Fax from the PSTN to this machine, the asterisk or diva or hylafax, does not detect this call as a fax and asterisk answer that call like a voice call. How sould I configure the software modem (iaxmodem) to use with the Eicon Diva card? What sould I do to make the Eicon Diva detects an incoming fax? If asterisk shall not accept this call, then configure asterisk not to do so. Either use another number, or check the transfercapability (if the sender did set this correct). Why don't you receive the fax via asterisk? You can answer the call and if a fax-tone is detected, you switch to receivefax. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Eicon Diva and incoming Fax
On Mon, 15 Oct 2007, voip crazy wrote: Dear Armin, the problem is my Eicon Diva Card does not detect aany fax-tone. Then the call is redirect as a voice call instead a fax call. How could I detect the fax.-tone with this kind of hardware? How could I enable receivefax? Are we talking about a DIVA Server BRI card? If yes, then the card can detect fax tone and you just need to enabled this in capi.conf. Then capicommand(receivefax|...) will help you. Armin 2007/10/15, Armin Schindler [EMAIL PROTECTED]: Hello VoipCrazy !? On Mon, 15 Oct 2007, voip crazy wrote: Hello all, I am trying to set up asterisk and hylafax to send and receibe fax. The machine is connected to the PSTN is an Eicon Diva BRI (1 BRI port). My problem is that , when I send a Fax from the PSTN to this machine, the asterisk or diva or hylafax, does not detect this call as a fax and asterisk answer that call like a voice call. How sould I configure the software modem (iaxmodem) to use with the Eicon Diva card? What sould I do to make the Eicon Diva detects an incoming fax? If asterisk shall not accept this call, then configure asterisk not to do so. Either use another number, or check the transfercapability (if the sender did set this correct). Why don't you receive the fax via asterisk? You can answer the call and if a fax-tone is detected, you switch to receivefax. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Eicon Diva and incoming Fax
On Mon, 15 Oct 2007, voip crazy wrote:
Dear Armin,
Bellow I send you my /etc/asterisk/capi.conf file, I just set
faxdetect=both, but the card isn`t detect an incoming fax call.
I use capicommand(receivefax|...), and work well, but I need that asterisk
or the diva card detects an incoming fax call to send it to a specific
context.
There are any way to use capicommand to detects fax incoming fax.
Don't set
softdtmf=on
relaxdtmf=on
because your DIVA card can do this with the onboard DSPs and fax detection
should work then too.
Armin
Thanks in advance.
VoipCrazy
--Capi.conf---
[general]
nationalprefix=0
internationalprefix=00
rxgain=1.0 ;linear receive gain (1.0 = no change)
txgain=1.0 ;linear transmit gain (1.0 = no change)
language=de ;set default language
;ulaw=yes;set this, if you live in u-law world instead of a-law
;jb. ;with Asterisk 1.4 you can configure jitterbuffer,
;see Asterisk documentation for all jb* setting available.
;mohinterpret=default ;Asterisk 1.4: default music on hold class when placed
on hold.
; interface sections ...
[ISDN] ;this example interface gets name 'ISDN1' and may be any
;name not starting with 'g' or 'contr'.
;Use one interface section for each isdn port!
;ntmode=yes ;if isdn card operates in nt mode, set this to yes
isdnmode=msn ;'MSN' (point-to-multipoint) or 'DID' (direct inward dial)
;when using NT-mode, 'DID' should be set in any case
incomingmsn=*;allow incoming calls to this list of MSNs/DIDs, * = any
;defaultcid=123 ;set a default caller id to that interface for dial-out,
;this caller id will be used when dial option 'd' is set.
;controller=0;ISDN4BSD default
;controller=7;ISDN4BSD USB default
controller=1 ;capi controller number of this interface/port
group=1 ;dialout group
;prefix=0;set a prefix to calling number on incoming calls
softdtmf=on ;enable/disable software dtmf detection, recommended for
AVM cards
relaxdtmf=on ;in addition to softdtmf, you can use relaxed dtmf
detection
faxdetect=both;enable faxdetection and redirection to EXTEN 'fax' for
incoming and/or
;outgoing calls. (default='off', possible values:
'incoming','outgoing','both')
accountcode=Canal-RDSI ;PBX accountcode to use in CDRs
;amaflags=default;AMA flags for CDR ('default', 'omit', 'billing', or
'documentation')
context=from-pstn ;context for incoming calls
;holdtype=hold ;when the PBX puts the call on hold, ISDN HOLD will be
used. If
;set to 'local' (default value), no hold is done and the
PBX may
;play MOH.
holdtype=local
;immediate=yes ;DID: immediate start of pbx with extension 's' if no
digits were
; received on incoming call (no destination number yet)
;MSN: start pbx on CONNECT_IND and don't wait for
SETUP/SENDING-COMPLETE.
; info like REDIRECTINGNUMBER may be lost, but this is
necessary for
; drivers/pbx/telco which does not send SETUP or
SENDING-COMPLETE.
;echosquelch=1 ;_VERY_PRIMITIVE_ echo suppression
;echocancel=yes ;EICON DIVA SERVER (CAPI) echo cancelation (yes=g165)
;(possible values: 'no', 'yes', 'force', 'g164', 'g165')
;echocancelold=yes;use facility selector 6 instead of correct 8 (necessary
for older eicon drivers)
;echotail=64 ;echo cancel tail setting (default=0 for maximum)
;echocancelnlp=1 ;activate non-linear-processing; this improves echo cancel
ratio, but might
;incorporate variable gain in the signal path.
bridge=yes ;native bridging (CAPI line interconnect) if available
;callgroup=1 ;PBX call group
;pickupgroup=1 ;PBX pickup group (which call groups are we allowed to
pickup)
;language=de ;set language for this device (overwrites default language)
;disallow=all;RTP codec selection (valid with Eicon DIVA Server only)
allow=all ;RTP codec selection (valid with Eicon DIVA Server only)
devices=2;number of concurrent calls (b-channels) on this controller
;(2 makes sense for single BRI, 30/23 for PRI/T1)
;jb. ;with Asterisk 1.4 you can configure jitterbuffer,
;see Asterisk documentation for all jb* setting available.
;mohinterpret=default ;Asterisk 1.4: default music on hold class when placed
on hold.
;qsig=on ;enable use of Q.SIG extensions.
--EOF--
2007/10/15, Armin Schindler [EMAIL PROTECTED]:
On Mon, 15 Oct 2007, voip crazy wrote:
Dear Armin,
the problem is my Eicon Diva Card does not detect aany fax-tone. Then
the
call is redirect as a voice call instead a fax call.
How could I detect the fax.-tone with this kind of hardware?
How could I enable receivefax?
Are we
Re: [asterisk-users] bridge on DIVA card and how to see it
On Thu, 6 Sep 2007, lemmel lemmel wrote: In the capi.conf file, there is a bridge option that allow to native bridging (CAPI line interconnect) if available, and I found this in the capi-user mailing list : I suggest you put bridge=yes into each interface. Then, when Asterisk bridges two channels, it looks for the possibility to do a native bridge (call the bridge code of the channel module). In case of SIP (when reinvite=yes is set), the SIP phones are set to send the voice data directly to the other phone and not bother Asterisk with that voice data. It is an interesting feature, and I put the right value in the conf file; but how to see the effect of this parameter ? All my test show that when asterisk run out of steam, the isdn calls too. Does this parameter function really ? How can I perform my test in order to ascertain it ? This function does work well. But it works if your ISDN card/driver supports it only. If you have a DIVA Server card, then you can use it. The bridge is done on the DIVA cards DSPs without CPU power. There are three possibilities to see if it really is working: 1) when you type 'capi show channels', you should see a 'G' (for bridGed) in the isdnstate column. 2) Use 'set verbose 5' and 'capi debug' to see the CAPI command when the call is activated. There should be some FACILITY_REQs and infos like 'Line Interconnect activated'. 3) Use 'set verbose 9' and 'capi debug' to see even all Voice Data as CAPI commands. If the bridge is active, there shouldn't be any DATA_B3 commands any more. Of course, this only works if both channels are CAPI and both controllers supports that. Also, if you have allow= set in your capi.conf (use of RTP with DIVA), the Line-Interconnect may not be activated (if so, please contact me). Armin ___ Sign up now for AstriCon 2007! September 25-28th. http://www.astricon.net/ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] bridge on DIVA card and how to see it
On Thu, 6 Sep 2007, lemmel lemmel wrote: Thanks for your quick answer :-). I am a rookie in all this telephony problem, so I'll try to be verbose. This function does work well. But it works if your ISDN card/driver supports it only. I currently have a Diva Server 4BRI Rev 2, and it seems that there is DSP on it (voice detection, and so on), so it is correct ?! Yes. This card supports it. There are three possibilities to see if it really is working: 1) when you type 'capi show channels', you should see a 'G' (for bridGed) in the isdnstate column. I just perform this check, and I didn't see the G. I have two S0 connected to a PBX (Siemens) and on the other side I have my Diva card. This is my capi.conf : * [contr1] isdnmode=did incomingmsn=* controller=1 group=1 relaxdtmf=on faxdetect=off accountcode= context=toto echocancelold=yes devices=4 bridge=yes * This is not quite correct. Your card actually has 4 controllers, so you need to create 4 sections (contr1, contr2, ...) with devices=2 each. !Oh, you use M-Adapter. Then your 4 channels should be working! Also, to make use of the DSPs, don't set softdtmf/relaxdtmf. And since you are using a newer driver, use echocancel=yes and leave echocancelold=off, otherwise your echo-canceler will not work. and I putted the 2 adapters to one M-Adapter. When I dialed 122 (on the first real adapter), I heard a voice asking for number to dialed, and I give the 103. When the call is established, I had : * Line-Name NTmode state i/o bproto isdnstate ton number contr1#04noConn I trans *BS 0x00 '107'-'122' contr1#03noConn O trans *BPS0x00 'b'-'103' contr1#02noDisc - trans 0x00 ''-'' contr1#01no- - trans 0x00 ''-'' * When I made a blind transfert to the 104, I had : Line-Name NTmode state i/o bproto isdnstate ton number * contr1#04noDisc - trans 0x00 ''-'' contr1#03noConn O trans *BPS0x00 'b'-'103' contr1#02noConn O trans *BPS0x00 'b'-'104' contr1#01no- - trans 0x00 ''-'' * Okay, both should be bridged. So somehow it is not activated. So please provide a log with set verbose 5 capi debug Also, if you have allow= set in your capi.conf (use of RTP with DIVA), the Line-Interconnect may not be activated (if so, please contact me). I didn't see this parameter in my capi.conf and in the capi.conf example, is it some specific DIVA parameter ? Where can I found those parameter, in Diva doc ? It is a capi.conf (chan-capi) parameter and is part of the example provided by chan-capi package. So far the DIVA Server cards are the only cards which can do RTP. But it is okay to leave it off. Armin ___ Sign up now for AstriCon 2007! September 25-28th. http://www.astricon.net/ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] bridge on DIVA card and how to see it
On Thu, 6 Sep 2007, lemmel lemmel wrote: So please provide a log with set verbose 5 capi debug Well, let's go for the debug log :-). In this capture, I used only one adapter, and I performed a call to asterisk, which dialed a number (as described in the previous mail) ; so this is the log : -- Executing [EMAIL PROTECTED]:2] Dial(CAPI/contr1#02/123-7, CAPI/contr1/b:103||tT) in new stack ... -- CAPI queue frame:[ TYPE: Control (4) SUBCLASS: Answer (4) ] [contr1#01] -- CAPI/contr1#01/103-8 answered CAPI/contr1#02/123-7 == contr1#02: Requested Indication-STOP for CAPI/contr1#02/123-7 CAPI devicestate requested for contr1#01/103 Hmm, it looks like there is something missing. Just after the 'answered' message, asterisk should say something like 'Attempting native bridge'. But since it doesn't appear, I think for some reason asterisk itself doesn't want to bridge here. Is this the complete log? Maybe you want to provide a full log (including call of the first channel and the hangup) to my personal mail? Armin ___ Sign up now for AstriCon 2007! September 25-28th. http://www.astricon.net/ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] bridge on DIVA card and how to see it
On Thu, 6 Sep 2007, lemmel lemmel wrote: Maybe you want to provide a full log (including call of the first channel and the hangup) to my personal mail? I just added an attachment to this mail. Thanks a lot :-) P.S.: this log was generated with verbosity to 5 and with capi debug Your Dial string has errors: CAPI/contr1/b:103||tT b: sets the caller number to 'b'. I think what you are trying to do is CAPI/contr1/103/b I'm not sure, but maybe tT lets asterisk avoiding the bridge. Armin ___ Sign up now for AstriCon 2007! September 25-28th. http://www.astricon.net/ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Dialogic support
On Tue, 21 Aug 2007, Wai Wu wrote: Can someone share pointers to Asterisk's Dialogic support? Which boards are supported, driver status, and etc. Which type of boards are you interested in? I don't know about other cards, but the DIVA Server ISDN cards are well supported. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] ISDN: Problems starting off
On Fri, 27 Jul 2007, Bertram Scharpf wrote: Hi, the first thing I did with Asterisk is listening to `demo-congrats' by Xlite on the same machine. This works perfectly. The config files are those shipped with the package. Now I want to listen to it over ISDN/Capi but I don't succeed. My `capi.conf' is like show in many tutorial on the web. In `extensions.conf' I just added the following lines: please provide your capi.conf. Which chan-capi version do you use? [capi-in] exten = 9876543,1,Goto(demo,1000,1) where 9876543 is my MSN without the area prefix. `demo' is the context that plays `demo-congrats'. The debug output I yield ends with (after a pause) DISCONNECT_IND ID=001 #0x0027 LEN=0014 Controller/PLCI/NCCI= 0x101 Reason = 0x0 DISCONNECT_RESP ID=001 #0x0027 LEN=0012 Controller/PLCI/NCCI= 0x101 -- ISDN1#02: DISCONNECT_IND on incoming without pbx, doing hangup. CAPI/ISDN1/9876543-2: set channel task to 1 == ISDN1#02: CAPI Hangingup for PLCI=0x101 in state 4 == ISDN1#02: Interface cleanup PLCI=0x101 CAPI devicestate requested for ISDN1/9876543 Seems that the MSN or even `capi-in' cannot be found at all. Yes, chan-capi seems to wait because of no match. Could anyone give me a hint what is going wrong here or at least what I have to diagnose next? The full debug log together with capi.conf should help. Armin ___ --Bandwidth and Colocation Provided by http://www.api-digital.com-- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan_capi install problems
On Thu, 31 May 2007, CSB wrote: On Sat, 26 May 2007, CSB wrote: I have installed Asterisk 1.2.18 am am trying to install chan_capi. The current RPM ftp://ftp.melware.net/Trixbox/chan_capi-1.0.1_1.2.14-1.i386.rpm installs but This precompiled RPM is for the previous trixbox asterisk version 1.2.14. A new RPM will follow soon... Do you have a rough idea of when? I hope to have time on the weekend. There is already a bug-report for this: http://bugs.melware.net/mantis/view.php?id=34 I had to wait, because the trixbox system didn't provide the correct headers. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Diva and Asterisk
On Thu, 31 May 2007, CSB wrote: I am trying to understand the difference between the divas4linux available from the Eicom/Dialogic website and the melware version. Am I right in thinking that the melware version is for Trixbox or Asterisk but the Dialogic version is for Asterisk only (i.e. will not work with Trixbox?). I've noticed that some of the things mentioned on the Dialogic web site as included with Divas e.g. web configuration, acopy2 are not available. Are they excluded from the melware version? This is partly correct. Let me try to explain: - both packages contain the same divas driver (they are based on the same sources and firmware) - the Eicon package is a RPM which patches your kernel for compliation of the drivers, the melware package just uses the kernel build-system, but doesn't modify your kernel. - the Eicon package includes additional drivers and files which are not used with the melware-package or because of closed-source nature. (E.g. web-conf: not used with melware; DIVA-TTY driver: closed source; ...) - Both packages can be used for asterisk, trixbox, etc. The difference is that the melware-package is also provided as precompiled RPM for trixbox. (Most people have problems compiling from source ;-) - melware package also contains the current version of the diva2i4l driver for kernel 2.6 (to use old isdn4linux devices with new divas driver). Also, when I try to reconfigure the Diva card I get the following message: Update CFGLib information ... failed ---DIVA CONFIGURATION: CFGLib DRIVER LOAD FAILED PLEASECHECK SYSTEM INSTALLATION (kernel version, missingfiles) ---DIVAS4LINUX SHUTDOWN OK.Is this because I'm using the melware divas4linux with Asterisk or is theresome other problem?Any advice is appreciated.RegardsCameron Which version is that? It should not happen again, when you do divas_stop.rc divas_cfg.rc and should be fixed in latest melware .tgz BTW: do you know that you can do conferencing with CAPI and non-CAPI channels with chan-capi and divas now? The onboard DSPs will be used for mixing the voice. I'm currently cloning/adapting app_meetme to work with chan-capi. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan_capi install problems
On Sun, 27 May 2007, CSB wrote: The current RPM ftp://ftp.melware.net/Trixbox/chan_capi-1.0.1_1.2.14-1.i386.rpm installs but This precompiled RPM is for the previous trixbox asterisk version 1.2.14. A new RPM will follow soon... I look forward to it. If you want to compile chan-capi by yourself, you need to install all dev- packages to have the needed header files. I think this should do it: yum -y install isdn4k-utils-devel asterisk-devel Having done that, I now get a message on asterisk startup: May 27 21:23:43 VERBOSE[4288] logger.c: [chan_capi.so]May 27 21:23:43 WARNING[4288] loader.c: /usr/lib/asterisk/modules/chan_capi.so: undefined symbol: ast_pickup_call May 27 21:23:43 WARNING[4288] loader.c: Loading module chan_capi.so failed! new chan-capi uses ast_pickup_call too. But this function is provided by module res_features. So you need to make sure to load res_features before chan-capi is loaded, e.g. in modules.conf: [modules] load=res_features.so load=chan_capi.so Armin But if the trixbox asterisk version again has special patches applied (something like jitterbuffer patch) which is not known to external modules like chan-capi, the compiled chan-capi may cause craches because it just doesn't match with the configured asterisk header files. I am intending to use Trixbox but in the meantime for testing purposes have installed Asterisk from source. Any further advice appreciated. Cameron ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan_capi install problems
On Sat, 26 May 2007, CSB wrote: I have installed Asterisk 1.2.18 am am trying to install chan_capi. The current RPM ftp://ftp.melware.net/Trixbox/chan_capi-1.0.1_1.2.14-1.i386.rpm installs but This precompiled RPM is for the previous trixbox asterisk version 1.2.14. A new RPM will follow soon... Asterisk dies on startup. The following appears in the log: May 27 03:28:18 asterisk1 kernel: divas: Diva Server V-4BRI-8 IRQ:7 SerNo:25290 May 27 03:28:18 asterisk1 kernel: divas: started with major 252 May 27 03:54:17 asterisk1 init: Trying to re-exec init The install notes say that the Asterisk version of the rpm must match so I guess that's the problem. exactly. Downloading and making ftp://ftp.melware.net/chan-capi/chan_capi-1.0.1.tar.gz gives me a bunch of errors mostly error: dereferencing pointer to incomplete type If you want to compile chan-capi by yourself, you need to install all dev- packages to have the needed header files. I think this should do it: yum -y install isdn4k-utils-devel asterisk-devel But if the trixbox asterisk version again has special patches applied (something like jitterbuffer patch) which is not known to external modules like chan-capi, the compiled chan-capi may cause craches because it just doesn't match with the configured asterisk header files. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Simultaneous Capacity
On Mon, 14 May 2007, Kapil Dhawan wrote: Just a quick brief I have a requirement of running 10 PRI's (300 Channels). I still have to decide on hardware and cards. Can you suggest some. As per my understanding it will be tough to go beyond 150. I didn't test exactly this yet, but from my experience it should work with the Dialogic DIVA Server 2 x 4PRI + 1 x 2PRI cards. Armin Alex Balashov wrote: On Mon, 14 May 2007, Kapil Dhawan said something to this effect: I want to try Asterisk with 10 PRI on a single Xeon machine with g711. Is it feasible. In truth, it is very unlikely. How are you planning to pick up the PRIs, anyway? 3 quad-span T1 cards? -- Alex Balashov [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Any other softPBX like Asterisk?
On Thu, 10 May 2007, Crazy Boy wrote: Hi Friends, Can anybody tell me other softPBX softwares like Asterisk? - OpenPBX - Freeswitch Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] Improve voice quality on Asterisk +chan_capi+DIVABRI
On Sat, 21 Apr 2007, Cosmin Prund wrote: Thanks a lot, that was it! I had both softdmtf=on and relasdtmf=on. I only touched softdmtf now, but I might have played with relaxdtfm before this. It now works fine with DTMF clamping activated. Okay, fine. That means you never used the onboard DSPs for DTMF detection before. Just as a note for all: with Eicon/Dialogic DIVA Server cards, the settings softdtmf and relaxdtmf don't make any sense. Armin Both logs don't show any DTMF activity. DMTF detection is not activated at all. Please make sure you DON'T have softdmtf=yes or relaxdtmf=yes in your capi.conf. -- Thanks, Cosmin Prund ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] Improve voice quality on Asterisk + chan_capi+DIVA BRI
On Fri, 20 Apr 2007, Cosmin Prund wrote: I've implemented my IVR using an FastAGI thing, using the READ application. core show application read shows no information on how the read function gets it's digits, I assume it does it the right way. With DTMF clamping off it works, with DTMF clamping on it no longer works. I've also toggled the softftfm setting in capi.conf, no luck ether way. Is there anything else I can try? Did I miss the obvious (it would not be my first) Can you please create a capi log: set verbose 5 capi debug to see what really happens via the interface? Armin -Original Message- From: [EMAIL PROTECTED] [mailto:asterisk-users- [EMAIL PROTECTED] On Behalf Of Armin Schindler Sent: 20 aprilie 2007 12:32 To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [asterisk-users] Improve voice quality on Asterisk + chan_capi+DIVA BRI On Fri, 20 Apr 2007, Cosmin Prund wrote: Ok, I've made all those changes, called my operator from an outside line and tried alternatively whispering / shouting into the mic, banging the microphone with a metal object and pressing DTMF digits. So far - so good, it seems to work. I've now got an other problem. Clamping DTMF disabled my IVR! Is there any way to enable/disable DTMF clamping on a per-call basis? Or better, disable DTMF only when the call makes it to an operator? This is possible, but such a command/feature must be implemented into chan-capi first. Anyway, even with DTMF clamping the DTMF detection is activated. So Asterisk should get the DTMF infos. Or is your IVR doing own DTMF detection on voice data? If yes, you should change that. Armin -Original Message- From: [EMAIL PROTECTED] [mailto:asterisk- users- [EMAIL PROTECTED] On Behalf Of Armin Schindler Sent: 19 aprilie 2007 14:35 To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [asterisk-users] Improve voice quality on Asterisk + chan_capi+ DIVA BRI On Thu, 19 Apr 2007, Cosmin Prund wrote: Hello everyone! I've got a Eicon Diva Server BRI card into my * box working just fine, but I wander if there's anything I can do to improve voice quality for my operators. I'm thinking something along the lines of auto gain and sudden noise suppression (like when you hit a fax machine or the other party accidently touches the dial pad on the phone). Does one of Asterisk, chan_capi or the Diva driver have support for such functionality? Sure, with the Dialogic (Eicon) DIVA Server card DSPs, you have the following possibilities: 1. Automatic Gain Control and Active Talker Evaluation in conference (by default automatically activated with three or more parties) 2. Recording Stream Automatic Gain Control 3. Manual Control of Signal Level 4. Manual control of the signal pitch and/or bitrate (rate conversion) 5. Suppression of DTMF tones. This feature can be activated using adapter configuration (for all calls) or on per call basis This is always good to activate this feature for operators to protect people from signals or in one gateway to prevent DTMF tones from passing through gateway in band. The DTMF tones are suppressed in the way which will not affect the quality of the voice signal in case voice signal and DTMF tones overlap. 6. Part 68 Voice Signal Limiter (Required in US, by default deactivated in Europe). This protects the ears from clicks and too loud signals. This feature can be activated using the configuration. This is good idea to activate Part 68 voice signal limiter to protect the people. This is the dynamic voice signal limiter in accordance with Part 68 of US requirements. The Part 68 Limiter, Audio Recording Automatic Gain Control (AGC of received signal) and the DTMF Clamping (Suppression of DTMF tones) are can be controlled using adapter configuration and do not require any change in the application (but can be controlled on the per call basis too, which is not implemented in chan-capi yet). Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing
RE: [asterisk-users] Improve voice quality on Asterisk + chan_capi+ DIVA BRI
On Fri, 20 Apr 2007, Cosmin Prund wrote: Ok, I've made all those changes, called my operator from an outside line and tried alternatively whispering / shouting into the mic, banging the microphone with a metal object and pressing DTMF digits. So far - so good, it seems to work. I've now got an other problem. Clamping DTMF disabled my IVR! Is there any way to enable/disable DTMF clamping on a per-call basis? Or better, disable DTMF only when the call makes it to an operator? This is possible, but such a command/feature must be implemented into chan-capi first. Anyway, even with DTMF clamping the DTMF detection is activated. So Asterisk should get the DTMF infos. Or is your IVR doing own DTMF detection on voice data? If yes, you should change that. Armin -Original Message- From: [EMAIL PROTECTED] [mailto:asterisk-users- [EMAIL PROTECTED] On Behalf Of Armin Schindler Sent: 19 aprilie 2007 14:35 To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [asterisk-users] Improve voice quality on Asterisk + chan_capi+ DIVA BRI On Thu, 19 Apr 2007, Cosmin Prund wrote: Hello everyone! I've got a Eicon Diva Server BRI card into my * box working just fine, but I wander if there's anything I can do to improve voice quality for my operators. I'm thinking something along the lines of auto gain and sudden noise suppression (like when you hit a fax machine or the other party accidently touches the dial pad on the phone). Does one of Asterisk, chan_capi or the Diva driver have support for such functionality? Sure, with the Dialogic (Eicon) DIVA Server card DSPs, you have the following possibilities: 1. Automatic Gain Control and Active Talker Evaluation in conference (by default automatically activated with three or more parties) 2. Recording Stream Automatic Gain Control 3. Manual Control of Signal Level 4. Manual control of the signal pitch and/or bitrate (rate conversion) 5. Suppression of DTMF tones. This feature can be activated using adapter configuration (for all calls) or on per call basis This is always good to activate this feature for operators to protect people from signals or in one gateway to prevent DTMF tones from passing through gateway in band. The DTMF tones are suppressed in the way which will not affect the quality of the voice signal in case voice signal and DTMF tones overlap. 6. Part 68 Voice Signal Limiter (Required in US, by default deactivated in Europe). This protects the ears from clicks and too loud signals. This feature can be activated using the configuration. This is good idea to activate Part 68 voice signal limiter to protect the people. This is the dynamic voice signal limiter in accordance with Part 68 of US requirements. The Part 68 Limiter, Audio Recording Automatic Gain Control (AGC of received signal) and the DTMF Clamping (Suppression of DTMF tones) are can be controlled using adapter configuration and do not require any change in the application (but can be controlled on the per call basis too, which is not implemented in chan-capi yet). Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] Improve voice quality on Asterisk +chan_capi+DIVA BRI
LLC = default AdditionalInfo BChannelinformation= default Keypadfacility = default Useruserdata = default Facilitydataarray = default CAPI devicestate requested for ISDN1/206364 -- CAPI/ISDN1/206364-29 Playing '/ram_sounds/intro-activare' (language 'de') CONNECT_ACTIVE_IND ID=001 #0x4a98 LEN=0015 Controller/PLCI/NCCI= 0x301 ConnectedNumber = default ConnectedSubaddress = default LLC = default CONNECT_ACTIVE_RESP ID=001 #0x4a98 LEN=0012 Controller/PLCI/NCCI= 0x301 CONNECT_B3_IND ID=001 #0x4a99 LEN=0013 Controller/PLCI/NCCI= 0x2c0301 NCPI= default CONNECT_B3_RESP ID=001 #0x4a99 LEN=0015 Controller/PLCI/NCCI= 0x2c0301 Reject = 0x0 NCPI= default CONNECT_B3_ACTIVE_IND ID=001 #0x4a9a LEN=0013 Controller/PLCI/NCCI= 0x2c0301 NCPI= default CONNECT_B3_ACTIVE_RESP ID=001 #0x4a9a LEN=0012 Controller/PLCI/NCCI= 0x2c0301 == ISDN1#02: Setting up echo canceller (PLCI=0x301, function=1, options=4, tail=0) FACILITY_REQ ID=001 #0x363c LEN=0024 Controller/PLCI/NCCI= 0x301 FacilitySelector= 0x8 FacilityRequestParameter= 01 00 06 04 00 00 00 00 00 FACILITY_CONF ID=001 #0x363c LEN=0022 Controller/PLCI/NCCI= 0x301 Info= 0x0 FacilitySelector= 0x8 FacilityConfirmationParameter = 01 00 02 00 00 -- ISDN1#02: Echo canceller successfully set up (PLCI=0x301) INFO_IND ID=001 #0x4bbc LEN=0017 Controller/PLCI/NCCI= 0x301 InfoNumber = 0x1e InfoElement = 82 88 INFO_RESP ID=001 #0x4bbc LEN=0012 Controller/PLCI/NCCI= 0x301 -- ISDN1#02: info element PI 82 88 ISDN1#02: In-band information available INFO_IND ID=001 #0x4bbd LEN=0017 Controller/PLCI/NCCI= 0x301 InfoNumber = 0x1e InfoElement = 82 83 INFO_RESP ID=001 #0x4bbd LEN=0012 Controller/PLCI/NCCI= 0x301 -- ISDN1#02: info element PI 82 83 ISDN1#02: Origination is non ISDN INFO_IND ID=001 #0x4bbe LEN=0017 Controller/PLCI/NCCI= 0x301 InfoNumber = 0x8 InfoElement = 80 90 INFO_RESP ID=001 #0x4bbe LEN=0012 Controller/PLCI/NCCI= 0x301 -- ISDN1#02: info element CAUSE 80 90 INFO_IND ID=001 #0x4bbf LEN=0015 Controller/PLCI/NCCI= 0x301 InfoNumber = 0x8045 InfoElement = default INFO_RESP ID=001 #0x4bbf LEN=0012 Controller/PLCI/NCCI= 0x301 -- ISDN1#02: info element DISCONNECT -- ISDN1#02: Disconnect case 3 -- CAPI queue frame: TYPE: Control (4) SUBCLASS: Hangup (1) ] [ISDN1#02] /CLI Output -- Thanks, Cosmin Prund -Original Message- From: [EMAIL PROTECTED] [mailto:asterisk-users- [EMAIL PROTECTED] On Behalf Of Armin Schindler Sent: 20 aprilie 2007 14:48 To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [asterisk-users] Improve voice quality on Asterisk +chan_capi+DIVA BRI On Fri, 20 Apr 2007, Cosmin Prund wrote: I've implemented my IVR using an FastAGI thing, using the READ application. core show application read shows no information on how the read function gets it's digits, I assume it does it the right way. With DTMF clamping off it works, with DTMF clamping on it no longer works. I've also toggled the softftfm setting in capi.conf, no luck ether way. Is there anything else I can try? Did I miss the obvious (it would not be my first) Can you please create a capi log: set verbose 5 capi debug to see what really happens via the interface? Armin -Original Message- From: [EMAIL PROTECTED] [mailto:asterisk- users- [EMAIL PROTECTED] On Behalf Of Armin Schindler Sent: 20 aprilie 2007 12:32 To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [asterisk-users] Improve voice quality on Asterisk + chan_capi+DIVA BRI On Fri, 20 Apr 2007, Cosmin Prund wrote: Ok, I've made all those changes, called my operator from an outside line and tried alternatively whispering / shouting into the mic, banging the microphone with a metal object and pressing DTMF digits. So far - so good, it seems to work. I've now got an other problem. Clamping DTMF disabled my IVR! Is there any way to enable/disable DTMF clamping on a per-call basis? Or better, disable DTMF only when
Re: [asterisk-users] Improve voice quality on Asterisk + chan_capi + DIVA BRI
On Thu, 19 Apr 2007, Cosmin Prund wrote: Hello everyone! I've got a Eicon Diva Server BRI card into my * box working just fine, but I wander if there's anything I can do to improve voice quality for my operators. I'm thinking something along the lines of auto gain and sudden noise suppression (like when you hit a fax machine or the other party accidently touches the dial pad on the phone). Does one of Asterisk, chan_capi or the Diva driver have support for such functionality? Sure, with the Dialogic (Eicon) DIVA Server card DSPs, you have the following possibilities: 1. Automatic Gain Control and Active Talker Evaluation in conference (by default automatically activated with three or more parties) 2. Recording Stream Automatic Gain Control 3. Manual Control of Signal Level 4. Manual control of the signal pitch and/or bitrate (rate conversion) 5. Suppression of DTMF tones. This feature can be activated using adapter configuration (for all calls) or on per call basis This is always good to activate this feature for operators to protect people from signals or in one gateway to prevent DTMF tones from passing through gateway in band. The DTMF tones are suppressed in the way which will not affect the quality of the voice signal in case voice signal and DTMF tones overlap. 6. Part 68 Voice Signal Limiter (Required in US, by default deactivated in Europe). This protects the ears from clicks and too loud signals. This feature can be activated using the configuration. This is good idea to activate Part 68 voice signal limiter to protect the people. This is the dynamic voice signal limiter in accordance with Part 68 of US requirements. The Part 68 Limiter, Audio Recording Automatic Gain Control (AGC of received signal) and the DTMF Clamping (Suppression of DTMF tones) are can be controlled using adapter configuration and do not require any change in the application (but can be controlled on the per call basis too, which is not implemented in chan-capi yet). Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Mon, 9 Apr 2007, Peer Oliver Schmidt wrote: Hello Armin (and happy easter), thanks for you continuing support. Can you please try HEAD version of SVN trunk (443)? Did checkout the 443. It works without any verbosity. THANK YOU! I'll buy you a beer, if you ever happen to come to the northern part of Germany. Thank you, but real thanks should go to the bug reporter of PR#28 on bugs.melware.net. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Tue, 3 Apr 2007, Armin Schindler wrote: On Tue, 3 Apr 2007, Peer Oliver Schmidt wrote: Hello Armin, thanks a lot for your help. Can you please do the same with 'showcapimsgs=2'? It may give more info on the commands itself, maybe some parameters are wrong here. Here you go. 17:23:17 is the magic time. This log below shows no error in parameters, but the problem is still the same: the fcpci driver doesn't respond and I cannot tell why. Can you please try HEAD version of SVN trunk (443)? It seems that the Fritz driver has a bug when registering at its CAPI interface. Armin Apr 3 17:23:09 server42 kernel: [263323.308388] fcpci: AVM FRITZ!Card PCI driver, revision 0.7.2 Apr 3 17:23:09 server42 kernel: [263323.308411] fcpci: (fcpci built on Feb 27 2007 at 21:22:25) Apr 3 17:23:09 server42 kernel: [263323.308421] fcpci: -- 32 bit CAPI driver -- Apr 3 17:23:10 server42 kernel: [263323.311559] PCI: Found IRQ 10 for device :00:0e.0 Apr 3 17:23:10 server42 kernel: [263323.311602] fcpci: AVM FRITZ!Card PCI found: port 0xdcc0, irq 10 Apr 3 17:23:10 server42 kernel: [263323.311613] fcpci: Loading... Apr 3 17:23:10 server42 kernel: [263323.311625] fcpci: Driver 'fcpci' attached to fcpci-stack. (152) Apr 3 17:23:10 server42 kernel: [263323.539987] fcpci: Stack version 3.11-07 Apr 3 17:23:10 server42 kernel: [263323.541140] kcapi: Controller 1: fcpci-dcc0-10 attached Apr 3 17:23:10 server42 kernel: [263323.541154] kcapi: card 1 fcpci-dcc0-10 ready. Apr 3 17:23:10 server42 kernel: [263323.541833] fcpci: Loaded. Apr 3 17:23:12 server42 kernel: [263325.975634] capi20: Rev 1.1.2.7: started up with major 68 (middleware+capifs) Apr 3 17:23:17 server42 kernel: [263330.892916] kcapi: put [0x1] FACILITY_REQ ID=001 #0x0001 LEN=0018 Apr 3 17:23:17 server42 kernel: [263330.892926] Controller/PLCI/NCCI = 0x1 Apr 3 17:23:17 server42 kernel: [263330.892933] FacilitySelector = 0x3 Apr 3 17:23:17 server42 kernel: [263330.892939] FacilityRequestParameter = 00 00 00 Apr 3 17:23:17 server42 kernel: [263330.892946] Apr 3 17:23:17 server42 kernel: [263330.893153] kcapi: got [0x1] FACILITY_CONF ID=001 #0x0001 LEN=0026 Apr 3 17:23:17 server42 kernel: [263330.893163] Controller/PLCI/NCCI = 0x1 Apr 3 17:23:17 server42 kernel: [263330.893169] Info= 0x0 Apr 3 17:23:17 server42 kernel: [263330.893176] FacilitySelector = 0x3 Apr 3 17:23:17 server42 kernel: [263330.893182] FacilityConfirmationParameter = 00 00 06 00 00\37703 00 00 Apr 3 17:23:17 server42 kernel: [263330.893190] Apr 3 17:23:17 server42 kernel: [263330.900689] kcapi: put [0x1] LISTEN_REQ ID=001 #0x0002 LEN=0026 Apr 3 17:23:17 server42 kernel: [263330.900699] Controller/PLCI/NCCI = 0x1 Apr 3 17:23:17 server42 kernel: [263330.900706] InfoMask= 0x Apr 3 17:23:17 server42 kernel: [263330.900713] CIPmask= 0x1fff03ff Apr 3 17:23:17 server42 kernel: [263330.900720] CIPmask2= 0x0 Apr 3 17:23:17 server42 kernel: [263330.900726] CallingPartyNumber = default Apr 3 17:23:17 server42 kernel: [263330.900733] CallingPartySubaddress = default Apr 3 17:23:17 server42 kernel: [263330.900739] -- Best regards Peer Oliver Schmidt PGP Key ID: 0x83E1C2EA ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Mon, 2 Apr 2007, Peer Oliver Schmidt wrote: Hello Armin, [EMAIL PROTECTED]:~# grep capi /var/log/asterisk/messages [Mar 31 16:17:03] ERROR[3850] chan_capi.c: Unable to listen on contr1 (error=0x100f) Is this helpful, or do you need more information? Yes, at this state it might be possible that less CPU power causes problems. The 'listen' command expects an answer and maybe it is coming too late. Can you please try the patch below? Index: chan_capi.c === --- chan_capi.c (revision 436) +++ chan_capi.c (working copy) @@ -631,7 +631,7 @@ error = LISTEN_CONF_INFO(CMSG); break; } - usleep(2); + usleep(10); tried the patch, but it did not work. It waits quite a long time before the chan-capi error message comes up, according to the time stamp it is about 12 seconds. It is kind of strange, that the whole startup process for asterisk usually takes only about 4-5 seconds. That's too long, normaly the confirmation message arrives within a few msecs. So it seems that the driver isn't responding. Do you need additional information? Which card/driver do you use? A debug log (capi trace) from the driver or kernelcapi helps to see what messages are wrong/missing. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Tue, 3 Apr 2007, Peer Oliver Schmidt wrote: Good Morning Armin, tried the patch, but it did not work. It waits quite a long time before the chan-capi error message comes up, according to the time stamp it is about 12 seconds. It is kind of strange, that the whole startup process for asterisk usually takes only about 4-5 seconds. That's too long, normaly the confirmation message arrives within a few msecs. So it seems that the driver isn't responding. Do you need additional information? Which card/driver do you use? [EMAIL PROTECTED]:~# lspci -s 0:0e -v 00:0e.0 Network controller: AVM Audiovisuelles MKTG Computer System GmbH A1 ISDN [Fritz] (rev 02) Subsystem: AVM Audiovisuelles MKTG Computer System GmbH FRITZ!Card ISDN Controller Flags: medium devsel, IRQ 10 Memory at ff001400 (32-bit, non-prefetchable) [size=32] I/O ports at dcc0 [size=32] [EMAIL PROTECTED]:~# capiinit status 1 fcpci running fcpci-dcc0-10A1 3.11-07 0xdcc0 10 Driver from ubuntu edgy I cannot tell anything about the AVM drivers. A debug log (capi trace) from the driver or kernelcapi helps to see what messages are wrong/missing. What is the best way to produce this? If the AVM driver can do that, I don't know. But on load of the module 'kernelcapi', you can specify the module parameter showcapimsgs=X where X is the verbose level. By default it is 0, which means no messges. You should set it to 3 to get the CAPI control messages on the kernel-console (logfile). Or even to 7 to have all CAPI messages (including data messages) which might be too much. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Tue, 3 Apr 2007, Peer Oliver Schmidt wrote: Hello Armin, here are the results, after modprobe kernelcapi showcapimsgs=3 /var/log/kern.log ... Apr 3 15:06:22 server42 kernel: [255120.102281] capi20: Rev 1.1.2.7: started up with major 68 (middleware+capifs) Apr 3 15:06:35 server42 kernel: [255133.725987] kcapi: appl 1 up Apr 3 15:06:35 server42 kernel: [255133.727235] kcapi: put [0x1] id#1 FACILITY_REQ len=18 Apr 3 15:06:35 server42 kernel: [255133.727712] kcapi: got [0x1] id#1 FACILITY_CONF len=26 Apr 3 15:06:35 server42 kernel: [255133.729933] kcapi: appl 1 down Apr 3 15:06:35 server42 kernel: [255133.730585] kcapi: appl 1 up Apr 3 15:06:35 server42 kernel: [255133.731478] kcapi: put [0x1] id#1 LISTEN_REQ len=26 Apr 3 15:06:52 server42 kernel: [255150.690252] kcapi: appl 1 down At 15:06:35 the loading stopped. Is this helpful, or do you need a higher verbosity? Well, it confirmes what I have expected, but I cannot tell why it happens. The driver doesn't respond to the LISTEN_REQ command, that's why chan-capi shows an error. So with this info, the driver is the problem. Can you please do the same with 'showcapimsgs=2'? It may give more info on the commands itself, maybe some parameters are wrong here. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Tue, 3 Apr 2007, Peer Oliver Schmidt wrote: Hello Armin, thanks a lot for your help. Can you please do the same with 'showcapimsgs=2'? It may give more info on the commands itself, maybe some parameters are wrong here. Here you go. 17:23:17 is the magic time. This log below shows no error in parameters, but the problem is still the same: the fcpci driver doesn't respond and I cannot tell why. Armin Apr 3 17:23:09 server42 kernel: [263323.308388] fcpci: AVM FRITZ!Card PCI driver, revision 0.7.2 Apr 3 17:23:09 server42 kernel: [263323.308411] fcpci: (fcpci built on Feb 27 2007 at 21:22:25) Apr 3 17:23:09 server42 kernel: [263323.308421] fcpci: -- 32 bit CAPI driver -- Apr 3 17:23:10 server42 kernel: [263323.311559] PCI: Found IRQ 10 for device :00:0e.0 Apr 3 17:23:10 server42 kernel: [263323.311602] fcpci: AVM FRITZ!Card PCI found: port 0xdcc0, irq 10 Apr 3 17:23:10 server42 kernel: [263323.311613] fcpci: Loading... Apr 3 17:23:10 server42 kernel: [263323.311625] fcpci: Driver 'fcpci' attached to fcpci-stack. (152) Apr 3 17:23:10 server42 kernel: [263323.539987] fcpci: Stack version 3.11-07 Apr 3 17:23:10 server42 kernel: [263323.541140] kcapi: Controller 1: fcpci-dcc0-10 attached Apr 3 17:23:10 server42 kernel: [263323.541154] kcapi: card 1 fcpci-dcc0-10 ready. Apr 3 17:23:10 server42 kernel: [263323.541833] fcpci: Loaded. Apr 3 17:23:12 server42 kernel: [263325.975634] capi20: Rev 1.1.2.7: started up with major 68 (middleware+capifs) Apr 3 17:23:17 server42 kernel: [263330.892916] kcapi: put [0x1] FACILITY_REQ ID=001 #0x0001 LEN=0018 Apr 3 17:23:17 server42 kernel: [263330.892926] Controller/PLCI/NCCI = 0x1 Apr 3 17:23:17 server42 kernel: [263330.892933] FacilitySelector = 0x3 Apr 3 17:23:17 server42 kernel: [263330.892939] FacilityRequestParameter = 00 00 00 Apr 3 17:23:17 server42 kernel: [263330.892946] Apr 3 17:23:17 server42 kernel: [263330.893153] kcapi: got [0x1] FACILITY_CONF ID=001 #0x0001 LEN=0026 Apr 3 17:23:17 server42 kernel: [263330.893163] Controller/PLCI/NCCI = 0x1 Apr 3 17:23:17 server42 kernel: [263330.893169] Info= 0x0 Apr 3 17:23:17 server42 kernel: [263330.893176] FacilitySelector = 0x3 Apr 3 17:23:17 server42 kernel: [263330.893182] FacilityConfirmationParameter = 00 00 06 00 00\37703 00 00 Apr 3 17:23:17 server42 kernel: [263330.893190] Apr 3 17:23:17 server42 kernel: [263330.900689] kcapi: put [0x1] LISTEN_REQ ID=001 #0x0002 LEN=0026 Apr 3 17:23:17 server42 kernel: [263330.900699] Controller/PLCI/NCCI = 0x1 Apr 3 17:23:17 server42 kernel: [263330.900706] InfoMask= 0x Apr 3 17:23:17 server42 kernel: [263330.900713] CIPmask= 0x1fff03ff Apr 3 17:23:17 server42 kernel: [263330.900720] CIPmask2= 0x0 Apr 3 17:23:17 server42 kernel: [263330.900726] CallingPartyNumber = default Apr 3 17:23:17 server42 kernel: [263330.900733] CallingPartySubaddress = default Apr 3 17:23:17 server42 kernel: [263330.900739] -- Best regards Peer Oliver Schmidt PGP Key ID: 0x83E1C2EA ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Mon, 2 Apr 2007, Peer Oliver Schmidt wrote: Hi, I have installed the above two, and have two questions: * Is there a reason (or better=a fix), why the chan-capi-1.0.0 does not compile together with Asterisk 1.4.2? Actually chan-capi-1.0.0 does compile with Asterisk 1.4.2. The problem is that after Asterisk 1.4.1 the asterisk version file was changed and the chan-capi script to detect the asterisk version does not recognize it. This problem is fixed in HEAD version of chan-capi, but you can also just change the line if grep -q ASTERISK_VERSION_NUM 0104 $INCLUDEDIR/version.h; then to if grep -q ASTERISK_VERSION_NUM .*104 $INCLUDEDIR/version.h; then in create_config.sh * Anyone else experiencing problems with chan-capi-HEAD not seeing the controller? If I run asterisk with verbose setting to 0, i.e. just asterisk -c chan-capi does not find the controller. Starting asterisk with verbosity turned up, most of the time the controller is found. Might this be a problem of missing CPU power (PII-400)? I don't know where this could happen. Do you have more info on what exactly is happening and when? Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] chan-capi-HEAD and Asterisk 1.4.2
On Mon, 2 Apr 2007, Peer Oliver Schmidt wrote: Hello Armin, Actually chan-capi-1.0.0 does compile with Asterisk 1.4.2. [..]but you [..] just change the line if grep -q ASTERISK_VERSION_NUM 0104 $INCLUDEDIR/version.h; then to if grep -q ASTERISK_VERSION_NUM .*104 $INCLUDEDIR/version.h; then in create_config.sh thanks, will try. * Anyone else experiencing problems with chan-capi-HEAD not seeing the controller? If I run asterisk with verbose setting to 0, i.e. just asterisk -c chan-capi does not find the controller. Starting asterisk with verbosity turned up, most of the time the controller is found. Might this be a problem of missing CPU power (PII-400)? I don't know where this could happen. Do you have more info on what exactly is happening and when? [EMAIL PROTECTED]:~# grep capi /var/log/asterisk/messages [Mar 31 16:17:03] ERROR[3850] chan_capi.c: Unable to listen on contr1 (error=0x100f) Is this helpful, or do you need more information? Yes, at this state it might be possible that less CPU power causes problems. The 'listen' command expects an answer and maybe it is coming too late. Can you please try the patch below? Armin -- Index: chan_capi.c === --- chan_capi.c (revision 436) +++ chan_capi.c (working copy) @@ -631,7 +631,7 @@ error = LISTEN_CONF_INFO(CMSG); break; } - usleep(2); + usleep(10); waitcount--; } if (!waitcount) ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] capi installation problem
I think you are mixing something here. The FritzCard is not a B1, so you don't need the b1 modules, the firmware and the /etc/capi.conf. You can either use the FritzCard driver (binary modules from AVM), or you use mISDN (which is also already loaded according to your lsmod). When using mISDN, you can either use the mISDN-CAPI to really provide a CAPI interface, or just don't use CAPI and use chan_misdn instead. Armin On Wed, 7 Mar 2007, Giedrius Augys wrote: Hello, I have problem with capi, I can't install it. I have putted all info what I did and what I get :). I think you can suggest me how to solve this problem.and thanking you in anticipation. I have ISDN Frtiz!Card PCI v2.1and I want to install it to my ubuntu box (kernel: 2.6.17-10-server). Using command lspci -vv , I can see that kernel finds this card: *02:0d.0 Network controller: AVM Audiovisuelles MKTG Computer System GmbH Fritz!PCI v2.0 ISDN (rev 02) Subsystem: AVM Audiovisuelles MKTG Computer System GmbH Fritz!PCI v2.0 ISDN Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium TAbort- TAbort- MAbort- SERR- PERR- Interrupt: pin A routed to IRQ 201 Region 0: Memory at ff9fec00 (32-bit, non-prefetchable) [size=32] Region 1: I/O ports at df80 [size=32] Capabilities: [40] Power Management version 2 Flags: PMEClk- DSI- D1- D2+ AuxCurrent=55mA PME(D0-,D1-,D2+,D3hot+,D3cold+) Status: D0 PME-Enable- DSel=0 DScale=0 PME-* I found tutorial in http://www.asteriskguru.com/tutorials/avm_b1.html .I have installed these packages: *ii capiutils 3.9.20060704-1 Utilities for CAPI-capable ISDN cards ii libcapi20-33.9.20060704-1 libraries for CAPI support ii libcapi20-dev 3.9.20060704-1 libraries for CAPI support ii avm-fritz-firmware-2.6.17-10 3.11+2.6.17.7-10.1 Firmware for AVM Fritz! ISDN hardware* and downloaded firmaware from * ftp://ftp.in-berlin.de/pub/capi4linux/firmware/b1/ .*My capi.conf is: *b1pci /usr/share/isdn/b1.t4 DSS1- - - - P2P* Then I exec command capiinit start, I have noting on output, but it load modules: [EMAIL PROTECTED]:~# lsmod Module Size Used by b1pci 10624 0 b1dma 17412 1 b1pci b1 25856 2 b1pci,b1dma capi 19392 0 kernelcapi 49664 4 b1pci,b1dma,b1,capi capifs 7176 2 capi ipv6 271136 12 lp 12964 0 mISDN_l2 44288 0 mISDN_l1 13192 0 avmfritz 25740 0 mISDN_isac 17280 1 avmfritz mISDN_core 75648 4 mISDN_l2,mISDN_l1,avmfritz,mISDN_isac* But when I execute command cappinfo, I get : [EMAIL PROTECTED]:~# capiinfo capi not installed - No such device or address (6). * ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Give Busy to the 3rd call on a BRI using chan_capi
What type of line is that? The 'base number' is also a MSN on lines I know. Or is it PtP with DID? Armin On Sat, 3 Feb 2007, Cosmin Prund wrote: Thanks, it really was easy. Unfortunately it only works for MSN's, not for the base number. Oh well, I'll just stop using the base number, I've got enough MSN's anyway. Thanks again. Armin Schindler wrote: On Thu, 1 Feb 2007, Cosmin Prund wrote: Any ideas? It should be simple... It is easy: read the README in chan-capi.org package ;-) Just look into the variable BCHANNELINFO and you will know if it is a call without b-channel (the third call). Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Give Busy to the 3rd call on a BRI using chan_capi
On Sat, 3 Feb 2007, Julian J. M. wrote: I'm still using asterisk 1.0.x bristuffed at one site.. Is there anything similar for this? When both channels are in use, 3rd call doesn't recive busy signal, but a message fromt he TelCo (something like The dialed number is not currently available). The asterisk version has nothing to do with that. Which chan-capi do you use? Armin On 2/3/07, Armin Schindler [EMAIL PROTECTED] wrote: What type of line is that? The 'base number' is also a MSN on lines I know. Or is it PtP with DID? Armin On Sat, 3 Feb 2007, Cosmin Prund wrote: Thanks, it really was easy. Unfortunately it only works for MSN's, not for the base number. Oh well, I'll just stop using the base number, I've got enough MSN's anyway. Thanks again. Armin Schindler wrote: On Thu, 1 Feb 2007, Cosmin Prund wrote: Any ideas? It should be simple... It is easy: read the README in chan-capi.org package ;-) Just look into the variable BCHANNELINFO and you will know if it is a call without b-channel (the third call). Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Give Busy to the 3rd call on a BRI using chan_capi
On Sat, 3 Feb 2007, Cosmin Prund wrote: Tzafrir Cohen wrote: Do you use Busy to send a bus signal to the other party? I use Busy. I have no idea how it works. When I call from my mobile phone to my PBX I get a busy signal and it seems I'm not being charged for the call (so it's not like * opened up the line and played the busy signal). It also works if I call from an other land line. If the call was not accepted yet and chan_capi receives the hangup command with cause busy from Asterisk, the call is 'rejected' with cause code busy. On PtMP connections this might not work, since there can be other devices which may accept the call. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Give Busy to the 3rd call on a BRI using chan_capi
On Thu, 1 Feb 2007, Cosmin Prund wrote: Any ideas? It should be simple... It is easy: read the README in chan-capi.org package ;-) Just look into the variable BCHANNELINFO and you will know if it is a call without b-channel (the third call). Armin Cosmin Prund wrote: Hello everyone: using chan_capi 0.7 and asterisk 1.4 Quick question: How can I detect the number of voice channels (B channels) in use at a given time. I'd like to call Busy if two B channels are used on my BRI to give the calling customer a Busy signal. Long question: On my single-line BRI (two channels) I'd like to give the 3rd simultaneous incoming call an busy signal. I already tested and the Busy function works very well (I've set up one of my MSN's to immediately call Busy). I also tested and I'm 100% sure the 3rd call makes it into the box while the other 2 channels are talking, so this is not a Telco problem and it can be fixed locally. Doing this on my side of the line (as opposed to having the Telco issue the Busy signal on my behalf) has an number of benefits: (a) I don't need to talk to the Telco (b) I *know* who called and I can call them back and (c) In a distant future I might use the capi channel's ability to transfer the call to a different POTS line since this doesn't use the B channel. Thanks, Cosmin Prund ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] About BRI / ISDN hardware. What to buy?
On Thu, 18 Jan 2007, Alberto Pastore wrote: I tested BRI-2M, 4BRI-8M, PRI-30M on several installations, even older 1.0 version cards (PCI 5v only) just work great. I use diva server drivers software source rpm from Eicon, chan_capi from www.melware.org (0.7.1) on asterisk 1.2.14 (kernel 2.6.17.3). We've deployed more than 40 PBX (from 1 bri to 8 bri) without a flaw. I'm only a little bit annoyed about not being able to take advantage of the onboard DSPs to perform audio transcoding, because of the lack of a suitable asterisk driver (the cards themselves support hardware gsm/g726 codecs, for instance). Sorry for that ;-) chan-capi already has rtp code to select one codec using also DSPs anti-jitter buffer. It is not fully tested and the full support for all codecs is still missing. Also, full conferencing using the DSPs is in progress. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] About BRI / ISDN hardware. What to buy?
On Thu, 18 Jan 2007, Cosmin Prund wrote: I finally found a price tag for the darn thing, at around 500 euros I can handle it. Qustion: Do they behave properly if I've got an other Digium TDM400 card in the system? How about installing two cards in the same server? At the moment I've only got 1 ISDN line plus a few analog lines going into the TDM but in the very near future we might want to get a second ISDN. The DIVA Server cards and its driver supports multiple cards. As long as other cards behave correctly on the bus, I don't see any problem here. Armin Alberto Pastore wrote: Jens Vagelpohl ha scritto: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 18 Jan 2007, at 18:31, Patrick wrote: I think http://www.melware.de carries the Eicon Server ISDN cards which have hardware echo cancellation. They are also the author of the chan_capi driver for Asterisk. I use the Eicon Server BRI cards with Asterisk myself and they work very well. I concur, I have a Eicon DIVA single port BRI card and it works very well. Cosmin, if you want to use it for Fax traffic as well make sure you do *not* get a V-BRI card. Those will not do Fax. jens Tried almost all cards (Junghanns, Sangoma, Beronet, some hfc-based oem cards, Eicon Diva Server). Eicon is expensive but is *REALLY* worth it. The other cards are just a waste of money (even if little money). If you want a reliable PBX (who doesn't want it?), Diva Server cards are the definitive choice. The best card ever. Zero echo problems, superb hardware echo cancellation. Top reliability. Excellent FAX support with Hylafax (only cards with builtin DSPs, that is, NOT the V-series, as pointed out by Jens). Easy driver installation and powerful utilities/configuration tools. I tested BRI-2M, 4BRI-8M, PRI-30M on several installations, even older 1.0 version cards (PCI 5v only) just work great. I use diva server drivers software source rpm from Eicon, chan_capi from www.melware.org (0.7.1) on asterisk 1.2.14 (kernel 2.6.17.3). We've deployed more than 40 PBX (from 1 bri to 8 bri) without a flaw. I'm only a little bit annoyed about not being able to take advantage of the onboard DSPs to perform audio transcoding, because of the lack of a suitable asterisk driver (the cards themselves support hardware gsm/g726 codecs, for instance). Alberto. ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] vzaphfc?
On Thu, 28 Dec 2006, Gavin Hamill wrote: On Thursday 28 December 2006 23:27, Tzafrir Cohen wrote: vzaphfc is not a complete replacement of bristuff. It replies on most of it. Rather, it replaces the zaphfc subdirectory with an improved ZapBRI driver for HFC-s-based PCI cards. Further, if you're looking for 'something else' re: cheapo ISDN cards, definately give Asterisk 1.4 and mISDN a look - no BRIStuff, no huge patches, no wacky stuff.. all Asterisk-core support that worked really well in the brief time I tested it. The key difference is rather than generating 8000 interrupts per second, the mISDN kernel driver (which itself can be thought of 'isdn4linux' version 2.0) mISDN is just one part of isdn4linux. don't forget the all other isdn drivers under linux and the core. If you talk about some 2.0 version, then you should say it is HiSax 2.0, because mISDN is the driver for the passive isdn cards like HiSax. Armin polls the card, leading to much lower system load, and no 'wanted 8 bytes, read 7!' errors from dmesg. Cheers, Gavin. ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] Diva Server V-BRI-2 and internal numbers
On Thu, 14 Dec 2006, Gregory Duchatelet wrote: It looks like 107 is busy ;-) Please increase verbosity, like set verbose 5 capi debug to see what is happening. Hi Armin, Verbose was at 30 :) 107 is not busy since i can call it from 102, which is another internal phone. All internal phones are busy for Asterisk... Here is the log with verbose at 100 and capi debug enabled : I should have seen that in your first email: your dialstring is wrong. You set 'b' as callerid for that call. I assume you want to have early-B3, so you should do Dial(CAPI/ISDN1/107/b) Armin -- Executing Dial(SIP/Greg-081f5a10, CAPI/ISDN1/b:107||rtT) in new stack data = ISDN1/b:107 parsed dialstring: 'ISDN1' 'b' '107' '' capi request for interface 'ISDN1' parsed dialstring: 'ISDN1' 'b' '107' '' == ISDN1: Call CAPI/ISDN1/107-1e (pres=0x00, ton=0x00) CONNECT_REQ ID=001 #0x1002 LEN=0047 Controller/PLCI/NCCI= 0x1 CIPValue= 0x1 CalledPartyNumber = 80107 CallingPartyNumber = 00 80b CalledPartySubaddress = default CallingPartySubaddress = default BProtocol B1protocol = 0x1 B2protocol = 0x1 B3protocol = 0x0 B1configuration= default B2configuration= default B3configuration= default GlobalConfiguration= default BC = default LLC = default HLC = default AdditionalInfo BChannelinformation= 00 00 Keypadfacility = default Useruserdata = default Facilitydataarray = default SendingComplete= default -- Called ISDN1/b:107 CAPI devicestate requested for ISDN1/107 CAPI devicestate requested for ISDN1/107 CONNECT_CONF ID=001 #0x1002 LEN=0014 Controller/PLCI/NCCI= 0x101 Info= 0x0 -- ISDN1: received CONNECT_CONF PLCI = 0x101 INFO_IND ID=001 #0x11a8 LEN=0016 Controller/PLCI/NCCI= 0x101 InfoNumber = 0x18 InfoElement = 8a INFO_RESP ID=001 #0x11a8 LEN=0012 Controller/PLCI/NCCI= 0x101 -- ISDN1: info element CHANNEL IDENTIFICATION 8a INFO_IND ID=001 #0x11a9 LEN=0015 Controller/PLCI/NCCI= 0x101 InfoNumber = 0x800d InfoElement = default INFO_RESP ID=001 #0x11a9 LEN=0012 Controller/PLCI/NCCI= 0x101 -- ISDN1: info element SETUP ACK INFO_IND ID=001 #0x11ab LEN=0015 Controller/PLCI/NCCI= 0x101 InfoNumber = 0x8002 InfoElement = default INFO_RESP ID=001 #0x11ab LEN=0012 Controller/PLCI/NCCI= 0x101 -- ISDN1: info element CALL PROCEEDING -- CAPI/ISDN1/107-1e is proceeding passing it to SIP/Greg-081f5a10 INFO_IND ID=001 #0x11ad LEN=0037 Controller/PLCI/NCCI= 0x101 InfoNumber = 0x1c InfoElement = 91 a1 13 02 02 8f 02 01 2200a a1 05003 02 01 00 82 01 01 INFO_RESP ID=001 #0x11ad LEN=0012 Controller/PLCI/NCCI= 0x101 -- ISDN1: info element FACILITY INFO_IND ID=001 #0x11ae LEN=0017 Controller/PLCI/NCCI= 0x101 InfoNumber = 0x8 InfoElement = 81 d8 INFO_RESP ID=001 #0x11ae LEN=0012 Controller/PLCI/NCCI= 0x101 -- ISDN1: info element CAUSE 81 d8 INFO_IND ID=001 #0x11af LEN=0015 Controller/PLCI/NCCI= 0x101 InfoNumber = 0x8045 InfoElement = default INFO_RESP ID=001 #0x11af LEN=0012 Controller/PLCI/NCCI= 0x101 -- ISDN1: info element DISCONNECT -- ISDN1: Disconnect case 1 -- CAPI/ISDN1/107-1e is busy == ISDN1: CAPI Hangingup -- ISDN1: activehangingup (cause=88) DISCONNECT_REQ ID=001 #0x1003 LEN=0018 Controller/PLCI/NCCI= 0x101 AdditionalInfo BChannelinformation= default Keypadfacility = default Useruserdata = default Facilitydataarray = default SendingComplete= default == Everyone is busy/congested at this time (1:1/0/0) -- Executing Hangup(SIP/Greg-081f5a10, ) in new stack == Spawn extension (macro-appel_sortant, s, 2) exited non-zero on 'SIP/Greg-081f5a10' in macro 'appel_sortant' == Spawn extension (macro-appel_sortant, s, 2) exited non-zero on 'SIP/Greg-081f5a10' CAPI devicestate requested for ISDN1/107 CAPI devicestate requested for ISDN1/107 DISCONNECT_CONF ID=001 #0x1003 LEN=0014 Controller/PLCI/NCCI
Re: [asterisk-users] Diva Server V-BRI-2 and internal numbers
On Wed, 13 Dec 2006, Gregory Duchatelet wrote: Hi, I have an asterisk with a DIVA Server V-BRI-2 card, connected to a Siemens PABX. From a SIP phone, I can call other internal SIP phones, external numbers (to PSTN), but I can't call internal phones connected to the internal phone network. When I call 107, which is an internal phone, heres the logs from asterisk: -- Executing Dial(SIP/Greg-081f5a10, CAPI/ISDN1/b:107||rtT) in new stack -- Called ISDN1/b:107 -- CAPI/ISDN1/107-1a is proceeding passing it to SIP/Greg-081f5a10 -- CAPI/ISDN1/107-1a is busy == ISDN1: CAPI Hangingup == Everyone is busy/congested at this time (1:1/0/0) -- Executing Hangup(SIP/Greg-081f5a10, ) in new stack == Spawn extension (macro-appel_sortant, s, 2) exited non-zero on 'SIP/Greg-081f5a10' in macro 'appel_sortant' == Spawn extension (macro-appel_sortant, s, 2) exited non-zero on 'SIP/Greg-081f5a10' BUT! If I call an internal isdn number like 122 which is a fax, the call is answered. How can I call 107 ? It looks like 107 is busy ;-) Please increase verbosity, like set verbose 5 capi debug to see what is happening. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] eicon diva BRI problems
On Thu, 7 Dec 2006, Klaus Darilion wrote: Hi (Armin?) ! Today I had a problem with Diva Server 4BRI-8M 2.0. Asterisk 1.2.12.1 chan_capi-cm-0.6.5 divas4linux-melware-3.0.f-106.622-1 Asterisk could not receive and make calls on the BRI ports, although the ports looked fine within Asterisk. I usually use /usr/lib/divas/divactrl dchannel -c 1 to test line activity. This time there was no activity (cryptic log messages) (I waited for 10 minutes). Then I restarted Asterisk - but no improvement. Then I wanted to remove and reload the diva kernel modules, but /usr/lib/divas/divas_stop.rc could not remove the modules. Also manual remove did not worked (module still in use). Then I rebooted the server and also updated to chan_capi-0.7.1 divas4linux-melware-3.0.5-106.702-1 as there were problem with removing the kernel modules and the Asterisk restart did not helped I suspect there is a bug in the diva kernel modules. Yes, such an error is caused by kernel modules. I searched for a changelog of the diva drivers, but couldn'T find them. Do you know of such a bug fixed in the newest version? No, I never had such problems (or received a report about that). Do you know of a changelog of the divas melware drivers? I don't have a changelog. If this problem appears again, please create a memory dump of the cards memory (divactrl can do that). This will help to find the problem, but the latest driver/firmware should be used. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] eicon diva BRI problems
On Thu, 7 Dec 2006, Klaus Darilion wrote: Armin Schindler wrote: I don't have a changelog. If this problem appears again, please create a memory dump of the cards memory (divactrl can do that). This will help to find the problem, but the latest driver/firmware should be used. Hi Armin! Can you please tell me exactly the proper statement to make this dump? I guess I want have much time to read the docs when problem happens again. divactrl ctrl -c 1 -File divadump.mem -CoreDump Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Error compiling Eicon Diva from source
On Thu, 7 Dec 2006, Matt Arnilo S. Baluyos (Mailing Lists) wrote: Hello everyone, I'm having some errors while compiling the source code for an Eicon Diva BRI card. I am using Fedora Core 6 with a compiled from source Linux kernel 2.6.19. I have downloaded the Eicon Diva source RPM (divas4linux_EICON-106.12-1.i386.rpm) and installed it already. When I go to /usr/lib/eicon/divas/src/ and run ./Build, everything starts correctly until that part where it has to do a make modules call. The last part of the divas.log file is as follows: quote HOSTCC scripts/conmakehash #+ LOG INFO: end modules_prepare cp: cannot stat `tmp.h': No such file or directory WARNING: Symbol version dump /usr/src/linux-2.6.19/Module.symvers is missing; modules will have no dependencies and modversions. CC [M] drivers/isdn/capi/kcapi.o CC [M] drivers/isdn/capi/capiutil.o drivers/isdn/capi/capiutil.c:20:26: error: linux/config.h: No such file or directory make[2]: *** [drivers/isdn/capi/capiutil.o] Error 1 make[1]: *** [drivers/isdn/capi] Error 2 make: *** [_module_drivers/isdn] Error 2 # + LOG INFO: pwd:/usr/lib/eicon/divas/src #! LOG ABORT EXECUTION DUE TO ERROR : Failed to call 'make modules' #! LOG ERROR INFO: make modules /quote Anyone know how I can get this working? Besides the fact that this question does not belong to this mailinglist, you should use the v3 divas driver from ftp.melware.net. This driver doesn't patch your kernel sources and compiles outside just using the headers. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] sip_write warning when executing Pickup of CAPI
On Tue, 5 Dec 2006, Tom Fanning wrote: I'm trying to pick up a ringing SIP phone (203) across the office with exten = *9,1,Pickup(783743) where 783743 is the local part of the number that our ISDN works on. I tried all of these first: exten = *9,1,Pickup(203) exten = *9,1,Pickup(SIP/203) exten = *9,1,Pickup([EMAIL PROTECTED]) and got a declined message back from my phone (snom 300), so I then switched to picking up the ringing ISDN line (it's BT ISDN2e on a pair of Eicon Diva BRI-2M cards) The Pickup(783743) works (the phone across the room stops ringing), but the calling party gets a nasty distorted noise back down the phone, and I get dozens of these messages: Dec 5 11:37:50 WARNING[26972]: chan_sip.c:2561 sip_write: Asked to transmit frame type 64, while native formats is 4 (read/write = 64/64) This sounds like a bug in Asterisk. chan-capi only accepts one format (alaw or ulaw), which is configured as native format. Asterisk here sends frames of type 64 (SLINEAR), which is wrong and not accepted by chan-capi. I cannot tell why Asterisk is doing that. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] Diva Server, chan_capi and tone detection
On Thu, 23 Nov 2006, Gregory Duchatelet wrote:
This would require a change in chan-capi. To get the extended tone
detection
indications, additional request/parameter via CAPI must be issued.
First, thanks for your reply.
Do you have the CxDtmf.pdf document, from Eicon ?
Yes.
If I understand good, you have to enable DTMF facilities 248, 249 and 250,
and then you receive DTMF code for tone detection :
0x81 for unidentified ton detected
0x80 for end of signal detected
0xC9 for human speech detected
Etc...
I didn't have a closer look into the values and commands yet, but basically
that should be right.
Another thing is, how do you want to get these indications for use in
your dialplan?
So, with DTMF code, you could handle it like for fax : redirect to extension
vad or something ...
That would mean to add for each of these signals an if {} to chan-capi
source. Not very nice, but will work.
Armin
___
--Bandwidth and Colocation provided by Easynews.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Diva Server, chan_capi and tone detection
On Tue, 21 Nov 2006, Gregory Duchatelet wrote: Hi all, I have a Diva Server V-BRI-2 card, which support, as written in reference guide: Extended tone processing (human talker detection, generation and detection of country-specific tones) I would like to detect human speech and fax tone with asterisk. I think that the diva card transmit a DTMF code when detecting voice, but chan_capi doesn't receive this DTMF code. I verbose it more, displaying all DTMF received, and only DTMF code CNG is received. Did you know how I can enable this detection (see DivaReportTones in Diva Server SDK) or how can I receive this DTMF in chan_capi ? This would require a change in chan-capi. To get the extended tone detection indications, additional request/parameter via CAPI must be issued. Another thing is, how do you want to get these indications for use in your dialplan? Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] capiAnswerFax
On Tue, 7 Nov 2006, Pedro Silva wrote:
Hello,
Anyone knows if chan_capi-0.7.1 includes the patch to support capiAnswerFax?
This patch is not added to chan-capi.org, but receivefax and sendfax is
available via capicommand(). Please see README of chan-capi 0.7.x package.
Armin
I tried to apply this patch (from http://www.mlkj.net/asterisk/) but
it give me errors...
Also i tried define one extension for fax receptions but this dont works:
exten = 1,1,Goto(handle_fax,s,1)
exten = fax,1,Goto(handle_fax,s,1)
[handle_fax]
exten = s,1,capiAnswerFax(/tmp/${UNIQUEID})
exten = s,2,Hangup()
With this code, a fax call to DID 1 must be attended and the fax
stored in /tmp, right?
This not works... :(
Thanks for any kind of possible help...
PS.
___
--Bandwidth and Colocation provided by Easynews.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
___
--Bandwidth and Colocation provided by Easynews.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Diva server 4bri and Portuguese BRI
On Wed, 1 Nov 2006, Pedro Silva wrote: Hello, The problem was wrong contexts defined like Marco said, and is solved. Now, i have another problem...of course :) On incoming calls, i only can receive calls if i define a line like the following, in extensions.conf: exten = _.,n,Dial(SIP/500,30,tr) (all incoming calls are redirected to extension 500). The problem is that i have some DDI's assigned by my telco (xxx302500 to xxx302509) and i need to route each DDI to diferent internal extension. If i define someting like exten = _0,n,Dial... (for DDI xxx302500) the call is not answered by asterisk. I think that asterisk cannot identify the destination DDI of the incoming call...is this normal? As you can see in the log below, the called number is just '0': CalledPartyNumber = 810 It seems DDI 0 of your line was called. So just do exten = 0,n,Dial... Armin This is the capi debug of one incoming call: asterisk1*CLI CONNECT_IND ID=001 #0x1975 LEN=0045 Controller/PLCI/NCCI= 0x401 CIPValue= 0x10 CalledPartyNumber = 810 CallingPartyNumber = 00 83X CalledPartySubaddress = default CallingPartySubaddress = default BC = 80 90 a3 LLC = default HLC = 91 81 AdditionalInfo BChannelinformation= default Keypadfacility = default Useruserdata = default Facilitydataarray = default -- CONNECT_IND (PLCI=0x401,DID=0,CID=X,CIP=0x10,CONTROLLER=0x1) ISDN1#02: msn='*' DNID='0' MSN == ISDN1#02: setting format alaw - 0x8 (alaw) == ISDN1#02: Incoming call 'X' - '0' INFO_IND ID=001 #0x1976 LEN=0017 Controller/PLCI/NCCI= 0x401 InfoNumber = 0x70 InfoElement = 810 INFO_RESP ID=001 #0x1976 LEN=0012 Controller/PLCI/NCCI= 0x401 -- ISDN1#02: info element CALLED PARTY NUMBER ISDN1#02: INFO_IND DID digits not used in this state. INFO_IND ID=001 #0x1977 LEN=0015 Controller/PLCI/NCCI= 0x401 InfoNumber = 0xa1 InfoElement = default INFO_RESP ID=001 #0x1977 LEN=0012 Controller/PLCI/NCCI= 0x401 -- ISDN1#02: info element Sending Complete CONNECT_RESP ID=001 #0x1977 LEN=0032 Controller/PLCI/NCCI= 0x401 Reject = 0x1 BProtocol B1protocol = 0x0 B2protocol = 0x0 B3protocol = 0x0 B1configuration= default B2configuration= default B3configuration= default ConnectedNumber = default ConnectedSubaddress = default LLC = default AdditionalInfo BChannelinformation= default Keypadfacility = default Useruserdata = default Facilitydataarray = default INFO_IND ID=001 #0x1978 LEN=0016 Controller/PLCI/NCCI= 0x401 InfoNumber = 0x18 InfoElement = 81 INFO_RESP ID=001 #0x1978 LEN=0012 Controller/PLCI/NCCI= 0x401 -- ISDN1#02: info element CHANNEL IDENTIFICATION 81 INFO_IND ID=001 #0x1979 LEN=0015 Controller/PLCI/NCCI= 0x401 InfoNumber = 0x8005 InfoElement = default INFO_RESP ID=001 #0x1979 LEN=0012 Controller/PLCI/NCCI= 0x401 -- ISDN1#02: info element SETUP ISDN1#02: IE SETUP / SENDING-COMPLETE already received. DISCONNECT_IND ID=001 #0x197b LEN=0014 Controller/PLCI/NCCI= 0x401 Reason = 0x0 DISCONNECT_RESP ID=001 #0x197b LEN=0012 Controller/PLCI/NCCI= 0x401 -- ISDN1#02: DISCONNECT_IND on incoming without pbx, doing hangup. CAPI/ISDN1/0-15: set channel task to 1 == ISDN1#02: CAPI Hangingup for PLCI=0x401 in state 4 == ISDN1#02: Interface cleanup PLCI=0x401 CAPI devicestate requested for ISDN1/0 Anyone can give me ideas about this problem? Thanks in advance! Best regards, PS. ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Diva server 4bri and Portuguese BRI
On Tue, 31 Oct 2006, Pedro Silva wrote: Hello, One problem is solved and another appears... :( I cannot receive incoming calls on trixbox. I defined one incoming route (any DID/any CID) and forwading these calls to a SIP extension. With capi and sip debug in asterisk -r console i dont detect any incoming activity... Did you use set verbose 5 capi debug ? If not, you should see anything there. But if you don't see activity with this verbose level too, this call is not signaled through capi. In that case you should create traces with divactrl ditrace (or the trace wizard) to get capi activity too. Armin In xlog console i have the following debug: 0:1898:127 - SIG-R(034) 08 01 0D 05 A1 04 03 80 90 A3 18 01 81 6C 0B 00 83 39 36 33 30 34 35 37 32 33 70 02 81 30 7D 02 91 81 Q.931 CR0d SETUP Sending complete Bearer Capability 80 90 a3 Channel Id 81 Calling Party Number 00 83 '963045723' Called Party Number 81 '0' HLC 91 81 0:1898:127 - SIG-S 0-6 e:805 0:1898:130 - CREATEID ok: context:1f assigned Id:3 freeIds=ec 0:1898:130 - alloc cr in use =4 0:1898:133 - SIG-X(008) 08 01 8D 45 08 02 80 95 Q.931 CR8d DISC Cause 80 95 'Call rejected' 0:1898:133 - SIG-x(008) 08 01 8D 5A 08 02 80 D8 Q.931 CR8d REL_COM Cause 80 d8 'Incompatible destination' 0:1898:133 - SIG-S 6-0 e:8c5 0:1898:134 - D-X(012) 00 01 14 16 08 01 8D 5A 08 02 80 D8 0:1898:135 - free cr in use =3 0:1898:135 - DELETEID ok: deleted Id:4 freeIds=ec 0:1898:155 - D-R(004) 00 01 01 16 So the problem appears to be Incompatible destination... but is problem in asterisk or is before asterisk, on diva card...? Tanks by any possible help! Best regards, PS. 2006/10/29, Pedro Silva [EMAIL PROTECTED]: Finally this works!!! :) Tanks to Alberto and Marco by your help! The problems are: - the cable was connected to the wong card port... :( - the card config needs to be: ETSI; TE; Point-to-Point (I thought that was point-to-multipoint). Best regards, PS. 2006/10/29, Pedro Silva [EMAIL PROTECTED]: Hello again Alberto! Anyway, to get more info, try to open a second shell and run /usr/lib/eicon/divas/xlog then on the first shell redo the telsampl test, then post the output of xlog off the list to my address (alberto at msoft-italia.com) This is the xlog output: 4:1736:074 - CREATEID ok: context:1f assigned Id:6 freeIds=eb 4:1736:074 - alloc cr in use =4 4:1736:076 - free cr in use =3 4:1736:077 - DELETEID ok: deleted Id:6 freeIds=eb 4:1736:078 - CREATEID ok: context:1f assigned Id:6 freeIds=eb 4:1736:078 - alloc cr in use =4 4:1736:080 - free cr in use =3 4:1736:080 - DELETEID ok: deleted Id:6 freeIds=eb 4:1736:081 - CREATEID ok: context:1f assigned Id:6 freeIds=eb 4:1736:081 - alloc cr in use =4 4:1736:083 - [1,0] dsp_assign 0016, 0, 2048 4:1736:083 - [1,0] CAI[00] 16 00 00 00 00 08 4:1736:084 - [1,0] Download 532 requested 4:1736:084 - MORE 4:1736:085 - SIG-X(029) 08 01 36 05 A1 04 03 80 90 A3 18 01 83 1E 02 80 83 70 0A 80 39 36 33 30 34 35 37 32 33 Q.931 CR36 SETUP Sending complete Bearer Capability 80 90 a3 Channel Id 83 Progress Indicator 80 83 Called Party Number 80 '963045723' 4:1736:085 - SIG-S 0-1 e:885 4:1736:087 - ACTIVATION_REQ 4:1744:147 - L1_DOWN 4:1744:150 - SIG-EVENT 08 4:1744:150 - SIG-EVENT 08 4:1744:150 - EVENT: Call failed in State 'Call initiated' Link disconnected, Layer-1 error (cable or NT) 4:1744:150 - SIG-S 1-0 e: 4:1744:151 - [1,0] dsp_release 4:1744:155 - free cr in use =3 4:1744:156 - DELETEID ok: deleted Id:6 freeIds=eb I disconnect the rj45 cable from alcatel pbx and connect that to the diva card (with alcatel pbx i can make calls normally). The green led of the diva card is activated when i connect the cable. So i dont understand why the message Link disconnected, Layer-1 error (cable or NT)... This debug is th same if the cable is connected to the NT or not. Any ideas...? Thanks! PS. ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] chan_capi and bristuff
On Fri, 27 Oct 2006, Michiel van Baak wrote: On 23:11, Thu 26 Oct 06, Armin Schindler wrote: snip/snip chan-capi-cm (chan-capi.org) is a complete rewritten version of chan-capi with more features and as far as I can tell, much more stable. You do faxing with chan-capi 0.3.5? But this isn't faxing via CAPI, right? As far as I know, chan-capi 0.3.5 does not support CAPI faxing. You are correct. chan-capi 0.3.5 does not support faxing out of the box. But there's a patch out there that fixes this. It will add CapiAnswerFax(). This function saves a .sff (structured fax file) stream to disk which you can run through sfftobmp and stuff. We create PDF files with some commandline tools and it's rock stable (2 years without a missed/corrupted fax on a system that takes like 10 to 15 faxes a week) Because it's working and we really believe in if it aint broken dont fix it we did not look into chan-capi-cm. This feature provides chan-capi.org for a long time and since version 0.7 you can send faxes too. offtopic My personal view on things: avoid PSTN/ISDN connections where possible and go with ITSP services. sangoma, quadbri, capi, tdm,... they all caused headaches where IAX simply works. Off course faxes wont work really good with ITSP but most of them have fax2email and email2fax. This is from a viewpoint of office PBX integration etc, not from ITSP viewpoint /offtopic I don't agree here. I have a few servers running with chan-capi.org and Eicon DIVA Server cards no problems at all. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Asterisk and ISDN and Hylafax
On Fri, 27 Oct 2006, Thomas Winter wrote: Am Thursday 26 October 2006 23:35 schrieben Sie: On Thu, 26 Oct 2006, Thomas Winter wrote: I would recommend the Eicon DIVA Server 4BRI cards. They have a capi interface which is used by chan-capi (chan-capi.org) and onboards DSPs for the faxing. You can use this for send and receive faxes and/or use capi4hylafax in parallel with asterisk/chan-capi. sounds good, you think it will run reliable? I do think so. I have this exactly this setup running twice. One setup even has 2 DIVA 4BRI cards, where the send one is running all ports in NT-mode to connect a legacy PBX. Regarding the 'parallel' hylafax, you just need to make sure that your setup is correct, e.g. asterisk should not be configured to accept calls which are meant for hylafax. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Asterisk and ISDN and Hylafax
On Fri, 27 Oct 2006, Olivier wrote: 2006/10/27, Armin Schindler [EMAIL PROTECTED]: On Fri, 27 Oct 2006, Thomas Winter wrote: Am Thursday 26 October 2006 23:35 schrieben Sie: On Thu, 26 Oct 2006, Thomas Winter wrote: I would recommend the Eicon DIVA Server 4BRI cards. They have a capi interface which is used by chan-capi (chan-capi.org) and onboards DSPs for the faxing. You can use this for send and receive faxes and/or use capi4hylafax in parallel with asterisk/chan-capi. sounds good, you think it will run reliable? I do think so. I have this exactly this setup running twice. One setup even has 2 DIVA 4BRI cards, where the send one is running all ports in NT-mode to connect a legacy PBX. Regarding the 'parallel' hylafax, you just need to make sure that your setup is correct, e.g. asterisk should not be configured to accept calls which are meant for hylafax. Armin What about telephony features using chan-capi and Asterisk ? Are those features on par with msidn+Asterisk or bristuff+Asterisk (maybe I'm mixing up things together) ? You can compare it with e.g. misdn+Asterisk. chan-capi is just another channel driver for Asterisk which provides access to ISDN/POTS hardware and drivers which support the CAPI interface. This includes - standard voice - DTMF - echo-cancel - Line-Interconnect - Fax - RTP ... Don't mix it with bristuff. As far as I know, bristuff consists of a) some additional zap driver. b) Asterisk changes/patches which alter asterisk features. [c) the old version of chan-capi] but actually it is not a channel-driver like chan-capi itself or chan-misdn. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] chan_capi and bristuff
On Thu, 26 Oct 2006, Michiel van Baak wrote: On 14:28, Thu 26 Oct 06, Tzafrir Cohen wrote: On Thu, Oct 26, 2006 at 10:48:52AM +0200, Michiel van Baak wrote: the chan_capi is merged with BRIStuff. This means you no longer have to download and compile chan_capi manually when you want to use a CAPI board with bristuffed asterisk. And while we're at it: How does the bristuff chan_capi compare the chan_capi-cm? I have no idea. I still use the patched chan_capi 0.3.5 and all I do with it is receiving faxes ;) 2 companies we installed the asterisk boxen for use chan_capi from bristuff 0.3.0 and they are happy with it (in combination with AVM Fritz!PCI cards using the avm driver) Since I have no experience with chan_capi-cm I cant tell you which one is better. chan-capi-cm (chan-capi.org) is a complete rewritten version of chan-capi with more features and as far as I can tell, much more stable. You do faxing with chan-capi 0.3.5? But this isn't faxing via CAPI, right? As far as I know, chan-capi 0.3.5 does not support CAPI faxing. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Problems with chan-capi and Eicon Diva 4BRI
On Mon, 23 Oct 2006, Klaus Darilion wrote: Hi! This weekend we had a problem with our Asterisk Box which ran flawlessly for nearly 4 weeks. The Asterisk server sits between the PSTN and a Siemens PBX and bridges 2 BRI lines. No calls, not incoming, not outgoing. The admin rebooted the Dell Box and then everything worked fine again. Now, I'm analyzing log files to find the cause. During the Asterisk outage the logfiles only show incoming (PSTN-Asterisk-PBX) calls, no outgoing. Thus I suspect that the Asterisk--PBX link was broken. In the Asterisk message file I only see Recovery on timer expiry errors, like below: ... Oct 20 17:18:26 VERBOSE[19772] logger.c: ISDN4#02: CAPI INFO 0x34e6: Recovery on timer expiry What does the timer expiry exactly mean? Was it a Layer2 or Layer 3 problem? Actually I don't know the reason for this, but it looks like it is coming from the connected device. How can I find out more or how can I activate more BRI debugging for the case it happens again? Use divactrl mlog -c X -o file.txt to get full log on port X. It also includes d-channel trace with description. Are there any known problems? We are using: Asterisk 1.2.12.1 chan_capi-0.7.0 divas4linux-melware-3.0.3-106.650-1 Diva Server 4BRI-8M 2.0 PCI No, this is mainly the same configuration I use for each installation these days. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Asterisk and ISDN and Hylafax
On Thu, 26 Oct 2006, Thomas Winter wrote: Hi, I have to set up an Asterisk with an 4-port BRI card. Hylafax should send and receive fax. Will this work reliable? Any recommandations for an 4-port BRI card? Other alternatives except analog fax units? I would recommend the Eicon DIVA Server 4BRI cards. They have a capi interface which is used by chan-capi (chan-capi.org) and onboards DSPs for the faxing. You can use this for send and receive faxes and/or use capi4hylafax in parallel with asterisk/chan-capi. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] CAPI channel not available but nobody is using the system
On Tue, 17 Oct 2006, Tim Sharp wrote: I have 23 CAPI channels defined and normally multiple channels are in use during the day for outbound calling. The problem is that every 3 or 4 months one of the channels becomes unavailable and then no calls can come in or go out on any of these channels. CAPI INFO shows Contr1: 23 B channels total, 22 B channels free. To fix the problem I reboot the asterisk server. First, is there a better way to reset the channels than rebooting? It depends where the problem really has its origin. If just asterisk (chan-capi) has a wrong channel count, it would be enough to unload chan-capi. Maybe asterisk itself need to be restarted. But if the real problem comes from the CAPI/ISDN driver, you need to reload these drivers. Which version of asterisk/chan-capi do you use? What ISDN hardware / CAPI driver do you use? Second, is there a way to bypass the unavailable channel in the dialplan? No. Third, what is causing the problem and can I prevent it? chan-capi counts the active channels when the CONNECT/DISCONNECT message of b-channels are indicated. If one of these messages are missing (it's a bug in the CAPI driver if that happens) the count is wrong. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] identifying Eicon Diva Server V-4BRI-8M vs 4BRI-8M
On Wed, 18 Oct 2006, Klaus Darilion wrote: Hi (Armin)! Does someone knows how to identify the type of the card? The delivery note says it is a V-4BRI-8M, whereas lspci reports a 4BRI-8M. What is it really? Are there any Eicon tools to identify the card type? As far as I know these cards are almost identical, but the PCI ID must be different. Maybe the pci id database doesn't have this difference... What PCI-ID does it have? 0xE012 = 4BRI-8M 0xE013 = 4BRI-8M V2 0xE016 = Voice 4BRI-8M 0xE017 = Voice 4BRI-8M V2 There is no special tool. When you load the divas driver, it should announce the cards found. And the divactrl utility uses divas to get the cards info and can tell the correct version as well, e.g.: divactrl ctrl -c 1 -CardName Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] CAPI channel not available but nobody is usingthe system
On Wed, 18 Oct 2006, Tim Sharp wrote: Armin, I am running 1.2.7.1 with an Eicon T1 board version 2 on Debian 2.4 I don't know the details on chan-capi / CAPI drivers. We did the install April of this year. How can I tell what I have? The divas driver version can be found in the syslog messages when the driver is loaded. I recommend to use the new V3 driver (ftp.melware.net). When you start asterisk (with verbosity 5) you can see the chan-capi messages including its version. It's an too old version if it is from April, please update, same ftp-server. Armin Thank you for your time. Tim -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Armin Schindler Sent: Wednesday, October 18, 2006 3:24 AM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: Re: [asterisk-users] CAPI channel not available but nobody is usingthe system On Tue, 17 Oct 2006, Tim Sharp wrote: I have 23 CAPI channels defined and normally multiple channels are in use during the day for outbound calling. The problem is that every 3 or 4 months one of the channels becomes unavailable and then no calls can come in or go out on any of these channels. CAPI INFO shows Contr1: 23 B channels total, 22 B channels free. To fix the problem I reboot the asterisk server. First, is there a better way to reset the channels than rebooting? It depends where the problem really has its origin. If just asterisk (chan-capi) has a wrong channel count, it would be enough to unload chan-capi. Maybe asterisk itself need to be restarted. But if the real problem comes from the CAPI/ISDN driver, you need to reload these drivers. Which version of asterisk/chan-capi do you use? What ISDN hardware / CAPI driver do you use? Second, is there a way to bypass the unavailable channel in the dialplan? No. Third, what is causing the problem and can I prevent it? chan-capi counts the active channels when the CONNECT/DISCONNECT message of b-channels are indicated. If one of these messages are missing (it's a bug in the CAPI driver if that happens) the count is wrong. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] core dump with 1.2.7.1 and chan-capi-cm 0.6.5
On Tue, 26 Sep 2006, Klaus Darilion wrote:
Hi Armin!
My Asterisk crashes once a day. The backtrace is:
Reading symbols from /usr/lib/asterisk/modules/func_enum.so...done.
Loaded symbols for /usr/lib/asterisk/modules/func_enum.so
Reading symbols from /usr/lib/asterisk/modules/func_uri.so...done.
Loaded symbols for /usr/lib/asterisk/modules/func_uri.so
Reading symbols from /lib/libnss_dns.so.2...done.
Loaded symbols for /lib/libnss_dns.so.2
#0 capi_handle_dtmf_fax (c=0x81b5b80) at chan_capi.c:1923
1923if (p-faxhandled) {
(gdb) bt
# 0 capi_handle_dtmf_fax (c=0x81b5b80) at chan_capi.c:1923
# 1 0x405df41f in capi_handle_facility_indication (CMSG=0xbd9ff944,
PLCI=258, NCCI=258, i=0x818a520) at chan_capi.c:2671
# 2 0x405dcb66 in capi_handle_msg (CMSG=0xbd9ff944) at chan_capi.c:3505
# 3 0x405da146 in do_monitor (data=0x0) at chan_capi.c:4186
# 4 0x40024e51 in pthread_start_thread () from /lib/libpthread.so.0
# 5 0x401ec8aa in clone () from /lib/libc.so.6
(gdb) quit
Is this is a known problem fixed in chan_capi-cm 0.7?
Yes, this is fixed in 0.7. At least it didn't show up again ;-)
Armin
___
--Bandwidth and Colocation provided by Easynews.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] asterisk 1.4 branch and chan-capi-0.7.0
On Tue, 26 Sep 2006, Guido Hecken wrote: Hi List, is there a known problem compiling chan-capi-0.7.0 against asterisk branch 1.4? chan-capi was not ported to Asterisk 1.4 yet. See bug http://bugs.melware.net/mantis/view.php?id=20 Armin System: Fedora Core 4 with Kernel 2.6.17-1.2142_FC4 AVM Fritz Card is present and fcpci running and up isdn4k-utils and isdn4k-utils-devel installed capi4hylafax installed make in chan_capi source said: gcc -pipe -fPIC -Wall -Wmissing-prototypes -Wmissing-declarations -g -I/usr/include -D_REENTRANT -D_GNU_SOURCE -O6 -march=i686 -Wno-missing-prototypes -Wno-missing-declarations -DCRYPTO -c -o chan_capi.o chan_capi.c chan_capi.c:146: warning: type defaults to 'int' in declaration of 'STANDARD_LOCAL_USER' chan_capi.c:146: warning: data definition has no type or storage class chan_capi.c:147: warning: type defaults to 'int' in declaration of 'LOCAL_USER_DECL' chan_capi.c:147: warning: data definition has no type or storage class chan_capi.c: In function 'capi_new': chan_capi.c:2078: error: 'struct ast_channel' has no member named 'type' chan_capi.c: In function 'pbx_capicommand_exec': chan_capi.c:4582: warning: implicit declaration of function 'LOCAL_USER_ADD' chan_capi.c:4597: warning: implicit declaration of function 'LOCAL_USER_REMOVE' chan_capi.c: At top level: chan_capi.c:5244: error: unknown field 'send_digit' specified in initializer chan_capi.c:5244: warning: initialization from incompatible pointer type make: *** [chan_capi.o] Error 1 Thanks for any hints and ideas Guido ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] asterisk 1.4 branch and chan-capi-0.7.0
On Tue, 26 Sep 2006, Guido Hecken wrote: -Ursprüngliche Nachricht- Von: Armin Schindler [mailto:[EMAIL PROTECTED] Gesendet: Dienstag, 26. September 2006 13:37 An: Asterisk Users Mailing List - Non-Commercial Discussion Betreff: Re: [asterisk-users] asterisk 1.4 branch and chan-capi-0.7.0 On Tue, 26 Sep 2006, Guido Hecken wrote: Hi List, is there a known problem compiling chan-capi-0.7.0 against asterisk branch 1.4? chan-capi was not ported to Asterisk 1.4 yet. See bug http://bugs.melware.net/mantis/view.php?id=20 Armin Armin, thanks for the info. Are there any plans on porting it to 1.4 and if yes, is there an approximate release date? Yes, of course it is planned ;-) But currently I have much to do, so if no one else has time, it will take a little bit longer and I cannot tell a date. Armin___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] asterisk / chan_capi problems
On Thu, 21 Sep 2006, Klaus Darilion wrote: Hi! I have problems with an Asterisk box which was running fine for some time but now causes problems (asterisk restarts, hangs ...). I use asterisk 1.2.7.1 with chan_capi-cm-0.6.5 and divas4linux-melware-3.0.f-106.622-1 In syslog I see lots of the following messages: Sep 20 07:02:41 ast01 kernel: kcapi: appl 1 ncci 0x130104 down Sep 20 07:02:43 ast01 kernel: kcapi: appl 1 ncci 0x60202 up Sep 20 07:02:43 ast01 kernel: kcapi: appl 1 ncci 0x10204 up Sep 20 07:03:12 ast01 kernel: kcapi: appl 1 ncci 0x10204 down Sep 20 07:03:12 ast01 kernel: kcapi: appl 1 ncci 0x60202 down Sep 20 07:06:25 ast01 kernel: kcapi: appl 1 ncci 0x20104 up Sep 20 07:06:40 ast01 kernel: kcapi: appl 1 ncci 0x20104 down Sep 20 07:07:13 ast01 kernel: kcapi: appl 1 ncci 0x70302 up What are the cause of this messages? May they be related with the asterisk crashes ? No, they are not related. These messages are just info messages from common kernelcapi driver about b-channel up: ncci 0x up b-channel down: ncci 0x down For the problems you have, some logs would be needed. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] asterisk / chan_capi problems
On Thu, 21 Sep 2006, Klaus Darilion wrote: Armin Schindler wrote: On Thu, 21 Sep 2006, Klaus Darilion wrote: Hi! I have problems with an Asterisk box which was running fine for some time but now causes problems (asterisk restarts, hangs ...). I use asterisk 1.2.7.1 with chan_capi-cm-0.6.5 and divas4linux-melware-3.0.f-106.622-1 In syslog I see lots of the following messages: Sep 20 07:02:41 ast01 kernel: kcapi: appl 1 ncci 0x130104 down Sep 20 07:02:43 ast01 kernel: kcapi: appl 1 ncci 0x60202 up Sep 20 07:02:43 ast01 kernel: kcapi: appl 1 ncci 0x10204 up Sep 20 07:03:12 ast01 kernel: kcapi: appl 1 ncci 0x10204 down Sep 20 07:03:12 ast01 kernel: kcapi: appl 1 ncci 0x60202 down Sep 20 07:06:25 ast01 kernel: kcapi: appl 1 ncci 0x20104 up Sep 20 07:06:40 ast01 kernel: kcapi: appl 1 ncci 0x20104 down Sep 20 07:07:13 ast01 kernel: kcapi: appl 1 ncci 0x70302 up What are the cause of this messages? May they be related with the asterisk crashes ? No, they are not related. These messages are just info messages from common kernelcapi driver about b-channel up: ncci 0x up b-channel down: ncci 0x down Just to make sure: What does it mean if a B channel goes up - is there a call started on this channel? No, the call already has started. The b-channel is the voice/data connection of the call. For the problems you have, some logs would be needed. No suspect logs at all. I've increase loglevel now and wait for new crashes. Okay. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Remote CAPI - ISDN over TCP/IP
Hi, remote CAPI already exists for a specific protocol (Bintec). This protocol is used by rcapidas well , a daemon which exports local CAPI ISDN hardware via TCP using the bintec protocol. Using rcapid you can use the ISDN hardware remotely within windows with the 'brickware' or from another linux host with the melware modified libcapi20 (ftp://ftp.melware.de). So you do can have an Asterisk server with chan-capi and the modified libcapi20 connected via TCP to the CAPI server running rcapid. If you want to use that Cisco router as CAPI server instead, then you just might need to extend the libcapi20 to support that protocol as well. Armin On Mon, 28 Aug 2006, Daniel Matos wrote: I posted about this a year or two ago: Currently we can find Cisco 801 ISDN Routers on ebay for about 40 euro. This equipment supports RCAPI (CAPI over TCP) Think of it as a isdn-ethernet adapter. There are other manufacturers who support this protocol in their ISDN Devices. (AVM, Bintec, Draytek) With the proper drivers you could give Asterisk an ISDN interface over a TCP/IP Connection. The problem is that altough the protocol exists for some years, there isn't any known implementation of this idea integrated with asterisk. One interesting use of this technology would be to be able to run Asterisk completely independent from hardware (in a virtual machine for example) while still supporting ISDN connectivity, using cheap, reliable and easy to find ISDN bridges (Cisco 801's for example) I know for sure that the protocol works, as I have some customers sending and receiving faxes, using voice mail and ISDN soft-phones on Windows, using RVS-COM. However I haven't seen any linux implementation, and no one talking about using it with asterisk. Today I have found some specifications and code samples from bintec (ftp://ftp.funkwerk-ec.com/pub/libcapi) that I hope someone with more knowledge than me can use to evaluate the possibility. Maybe we can start a bounty? I would gladly pay a reasonable amount for a working asterisk-Cisco 801 solution. Best Regards, Daniel Matos ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [asterisk-users] Remote CAPI - ISDN over TCP/IP
On Mon, 28 Aug 2006, Daniel Matos wrote: remote CAPI already exists for a specific protocol (Bintec). This protocol is used by rcapidas well , a daemon which exports local CAPI ISDN hardware via TCP using the bintec protocol. Using rcapid you can use the ISDN hardware remotely within windows with the 'brickware' or from another linux host with the melware modified libcapi20 (ftp://ftp.melware.de). So you do can have an Asterisk server with chan-capi and the modified libcapi20 connected via TCP to the CAPI server running rcapid. If you want to use that Cisco router as CAPI server instead, then you just might need to extend the libcapi20 to support that protocol as well. Armin Does anybody know if the Cisco implementation is different from Bintec? I cannot tell for sure, but I doubt that since the Bintec protocol is proprietary. Historically which one came first? I would be very happy to know that Cisco used the (apparently) open protocol from Bintec to enhance features, while keeping compatibility... The bintec protocol is not open (as far as I know). For rcapid it was re-engineered. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Calls over VPN
On Tue, 22 Aug 2006, Joseph wrote: I was thinking of using openVPN No problem. We are using it without problems. Armin -- #Joseph On Tue, 2006-08-22 at 22:06 -0500, Brandon Galbraith wrote: MPLS is a VPN, but it doesn't use encryption in most cases. -brandon On 8/22/06, Paul Hales [EMAIL PROTECTED] wrote: We did a setup of 70 sites connected back to a central Asterisk box, and it worked very well over an MPLS VPN. ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] Calls over VPN
Yes, I agree. And one more thing: With some encrypt setups of openvpn the data path is 'on hold' when openvpn recreates/renegotiates a new encryption. This means that you have a short interrupt (some milliseconds) when openvpn server a) establishes a new connection b) re-creates the encrypt-key (e.g. once per hour) To avoid a) I setup one openvpn server instance for each (important) VPN connection. To reduce b) I increased the reneg cycle. Armin On Wed, 23 Aug 2006, Simon Woodhead wrote: We've done this with OpenVPN and it works fine. I'd recommend that the VPN server is not on the same box as Asterisk. Stick it on a firewall/gateway box giving access to the network containing the Asterisk boxes behind it. This way the Asterisk box(es) is seeing normal unencrypted traffic and the VPN server(s) can be specified to meet the VPN requirement. On 8/23/06, Joseph [EMAIL PROTECTED] wrote: Is anybody making calls over VPN? If so what is the penalty as encryption is involved. I was planning to use VPN to register Sipura units to my local asterisk this way I don't have to deal with NAT issues. -- #Joseph ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] New people in this world and his problem with ISDN
On Mon, 7 Aug 2006, Dominik KieÅb wrote: Hi all, I'm new in Asterisk world, but it's very interest for me. I have some experience in CC and CTI. Now, is time for me on Asterisk. I use environment: Fedora Core 5 Asterisk 1.2 Eicona BRA Cards Chan_capi X-Lite for SIP users and HiPath 3550 for PSTN users I test different configuration and possibilities, and I must talk: It's great. But, off course I have problem. ISDN BRA connected to Eicona card has number 115. When I was try make two connection to this port and route this to SIP user or put to queue, all was great, but when I try transfer one PBX user (ext. 102) to another (ext. 101) it not works. Off course I want use for this Asterisk, make call from 101 to asterisk and make transfer to another I have error: Your call from 102 to 101 is not accepted by your ISDN PBX with the following error: CAPI INFO 0x34bf: Service or option not available, unspecified This is not a problem of Asterisk/chan-capi/ISDN-card. Seems to be the setup of the PBX. Armin ___ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
