Re: [asterisk-users] Using Asterisk 10.6 as a T38 Fax gateway
I forgot to ask: Do I have to load "res_fax" or "app_fax" to use the T38 gateway capability? -- _ -- 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
[asterisk-users] Using Asterisk 10.6 as a T38 Fax gateway
Hi all, and thanks for taking the time to read this.
I am trying to configure Asterisk 10.6 as a T38 Fax gateway. I am
receiving calls through the PSTN and want to send them to my VoIP
carriers as T38. This is my dialplan:
[fax]
exten => _X.,1,Set(FAXOPT(t38gateway)=yes,20)
exten => _X.,n,Dial(SIP/${EXTEN}@x.x.x.x)
I have tried with both FAXOPT(t38gateway) and FAXOPT(gateway). I have
also tried setting t38pt_udptl = yes,redundancy in sip.conf. None of
these things work. When we send a fax:
1. Asterisk does NOT send a REINVITE with the t38 offered. Reading the
documentation, it should detect the fax tone with the audiohook and
then send a REINVITE with t38 capability.
2. Asterisk does not offer t38 in the SDP of the initial INVITE. This
is not a problem if it correctly detects and REINVITES for faxes, but
our destination carriers tell us that they cannot do the REINVITE
themselves because we do not offer t38 in our SDP, so they believe we
do not have that capability.
Obviously I would prefer to just detect the fax myself and have
asterisk do the REINVITE.
I have read all of the documentation on the asterisk wiki (which is
rather short) and anything else I could find online. Unfortunately
most of it is out of date and refers to asterisk versions 1.4 to 1.8,
which do not have T38 Gateway capability.
Does anybody have any experience in making this work?
Thank you!
Alex
--
_
-- 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
[asterisk-users] FastAGI script and DIAL execution
Hi all,
I am trying to control the whole call using a FastAGI script. To that
effect I launch a FastAGI script (written with asterisk-java).
Basically, I want to DIAL from within the FastAGI script. When the call
ends I want to control the hangup (if executed at the remote end), and
depending on the cause, dial again, play a message, or hang up. This is a
pretty standard telephony scenario. I did it before by executing the AGI,
setting variables, calling the DIAL command from the dialplan, and then
executing a second AGI script for the cleanup logic. However, now that I am
using FastAGI it seems like a better idea to keep the AGI script alive
during the duration of the call. This gives me a lot of control and
fexibility on reporting.
However, as far as I can tell, once the called party hangs up, the CDR is
generated and posted, _even though my script is still in execution_! As you
can see from the sample below, the called party hangs up, and dialplan
execution starts immediately at the h extension, even though my script is
still running. In fact, I have quite a bit of cleanup to do, adding
variables to the CDR's, and none of them are saved! I believe this is
because the CDR is already finised.
It's like if once you call the DIAL aplication, the dialplan forks off and
your script is running in a different place. I do not understand it. I
assumed when I called DIAL from within a script, that the script execution
would suspend, but be resumed once the DIAL command returned, but this is
not what is happening.
Is there any way to get that behaviour?
Regards,
Alex
"Entering customer extension"
-- Executing [62999@customer:2] Verbose("SIP/139255423-004c",
"5,"Dialed - 62999"") in new stack
> "Dialed - 62999"
-- Executing [62999@customer:3] Set("SIP/139255423-004c",
"origincontext=customer") in new stack
-- Executing [62999@customer:4] Goto("SIP/139255423-004c",
"transform,62999,1") in new stack
-- Goto (transform,62999,1)
-- Executing [62999@transform:1] Goto("SIP/139255423-004c",
"customer,003462999,transform") in new stack
-- Goto (customer,003462999,5)
-- Executing [003462999@customer:5]
Verbose("SIP/139255423-004c", "5,"New dialnum - 003462999"") in new
stack
> "New dialnum - 003462999"
-- Executing [003462999@customer:6] Set("SIP/139255423-004c",
"CDR(server)=7") in new stack
-- Executing [003462999@customer:7] Set("SIP/139255423-004c",
"CDR(srcip)=") in new stack
-- Executing [003462999@customer:8] AGI("SIP/139255423-004c",
"agi://localhost/auth") in new stack
AGI Tx >> agi_network: yes
AGI Tx >> agi_network_script: auth
AGI Tx >> agi_request: agi://localhost/auth
AGI Tx >> agi_channel: SIP/139255423-004c
AGI Tx >> agi_language: es
AGI Tx >> agi_type: SIP
AGI Tx >> agi_uniqueid: 1340616655.76
AGI Tx >> agi_version: 10.5.0
AGI Tx >> agi_callerid: 139255
AGI Tx >> agi_calleridname: unknown
AGI Tx >> agi_callingpres: 0
AGI Tx >> agi_callingani2: 0
AGI Tx >> agi_callington: 0
AGI Tx >> agi_callingtns: 0
AGI Tx >> agi_dnid: 62999
AGI Tx >> agi_rdnis: unknown
AGI Tx >> agi_context: customer
AGI Tx >> agi_extension: 003462999
AGI Tx >> agi_priority: 8
AGI Tx >> agi_enhanced: 0.0
AGI Tx >> agi_accountcode: 704741
AGI Tx >> agi_threadid: 1104279872
AGI Tx >>
AGI Rx << GET VARIABLE "CDR(src)"
AGI Tx >> 200 result=1 (139255423)
AGI Rx << SET VARIABLE "CDR(accountcode)" "704741"
AGI Tx >> 200 result=1
AGI Rx << SET VARIABLE "CDR(dest_id)" "507"
AGI Tx >> 200 result=1
AGI Rx << SET VARIABLE "CDR(routeplan)" "11261"
AGI Tx >> 200 result=1
AGI Rx << SET VARIABLE "CDR(carrier)" "69"
AGI Tx >> 200 result=1
AGI Rx << EXEC "Dial" "SIP/10003462999@x.x.x.x"
-- AGI Script Executing Application: (Dial) Options:
(SIP/10003462999@x.x.x.x)
== Using SIP RTP CoS mark 5
-- Called SIP/10003462999@193.17.66.71
-- SIP/193.17.66.71-004d is making progress passing it to
SIP/139255423-004c
-- SIP/193.17.66.71-004d is ringing
-- SIP/193.17.66.71-004d is making progress passing it to
SIP/139255423-004c
-- SIP/193.17.66.71-004d answered SIP/139255423-004c
-- Executing [h@customer:1] Set("SIP/139255423-004c",
"CDR(q931)=16") in new stack
-- Executing [h@customer:2] Set("SIP/139255423-004c",
"CDR(userfield)={"agi":"","a-leg-id":"2118d872-305e-4bb4-8c47-30e1514cb934","b-leg-id":"36b232e73ac326bd0407b1594627c589@y.y.y.y:5060"}")
in new stack
AGI Tx >> 200 result=-1
AGI Tx >> HANGUP
AGI Rx << GET VARIABLE "HANGUPCAUSE"
AGI Tx >> 200 result=1 (16)
AGI Rx << GET VARIABLE "Q16"
AGI Tx >> 200 result=1 (0)
AGI Rx << SET VARIABLE "AJ_AGISTATUS" "SUCCESS"
AGI Tx >> 200 result=1
-- AGI Script agi://localhost/auth completed,
returning 4
AGI Tx >> HANGUP
== Spawn extension (customer, 003462999, 8) exited non-zero on
'SIP/139255423-004c'
--
_
-- Bandwid
Re: [asterisk-users] Receiving musinc on hold instead of ring
Hi Tarek, Yes, after running some more detailed packet captures, it seems that the SDP sent has the sendonly media attribute. I do not know if it is the Sonus switch, but the problem is identical to yours. Unfortunately setting canreinvite=yes for that peer does not solve the problem. I am guessing this is because the other leg of the call has canreinvite=no. This is necessary for correct billing. Should I submit it as an asterisk bug? Is there something else I can try to fix this interconnection? Thanks for your help! Alex On Wed, Sep 28, 2011 at 7:34 PM, Tarek Sawah wrote: > i have faced this problem with one of the major VoIP whole providers in > India .. they have a new platform with Sonus switches.. which does not > support sendrecv media attribute .. however a work around that may work for > you .. is enabling re-invite on their peer. > let me know if this works out for you. > > > Tarek Sawah > > Information Technology Adviser > > Integrated Digital Systems > > CCNP, MCSE, RHCE, TELECOM > > USA: +1 386 492 9993 > > > >> From: alexreca...@gmail.com >> Date: Wed, 28 Sep 2011 18:59:39 +0200 >> To: asterisk-users@lists.digium.com >> Subject: Re: [asterisk-users] Receiving musinc on hold instead of ring >> >> > this is related to your carrier's SIP messages as they are sending a >> > sendonly attribute instead of sendrecv (taking a wild guess here) your >> > asterisk will act as if the call was placed on hold thus the MOH butts >> > in. >> > an sip debug log for a similar call will be more helpful? >> >> Thanks for the answer Tarek! I will try to obtain a full SIP trace >> tonight. If the problem is indeed that the carrier is sending the >> sendonly attribute in the SDP instead of sendrecv, what can I do? Is >> there anything I can configure on my side? >> >> Thanks again, >> >> Alex >> >> -- >> _ >> -- 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 > > -- > _ > -- 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 > -- _ -- 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] Receiving musinc on hold instead of ring
> this is related to your carrier's SIP messages as they are sending a > sendonly attribute instead of sendrecv (taking a wild guess here) your > asterisk will act as if the call was placed on hold thus the MOH butts in. > an sip debug log for a similar call will be more helpful? Thanks for the answer Tarek! I will try to obtain a full SIP trace tonight. If the problem is indeed that the carrier is sending the sendonly attribute in the SDP instead of sendrecv, what can I do? Is there anything I can configure on my side? Thanks again, Alex -- _ -- 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
[asterisk-users] Receiving musinc on hold instead of ring
Hi all and thanks for reading.
I am having a very strange issue. When dialing out with a certain
carrier, asterisk 1.6.20 will play music on hold instead of a ring
tone, although this behaviour is NOT what I want.
Example dialplan execution:
-- Executing [0021266xxx@test:13] Progress("SIP/100-1e04", "") in new stack
-- Executing [0021266xxx@test:14]
Dial("SIP/100-1e04","SIP/21266xxx@x.x.x.x") in new stack
-- Called 21266xxx@x.x.x.x
-- Call on SIP/x.x.x.x-1e05 placed on hold
-- Started music on hold, class 'default', on SIP/100-1e04
-- SIP/x.x.x.x-1e05 is making progress passing it to SIP/100-1e04
Now, a SIP packet capture shows no trace of the call being put on hold!
Sample wireshark capture for the same call:
x.x.x.x -> y.y.y.y SIP/SDP Request: INVITE sip:21266xxx@x.x.x.x, with
session description
y.y.y.y -> x.x.x.x SIP Status: 100 Giving a try
y.y.y.y -> x.x.x.x SIP/SDP Status: 180 Ringing, with session description
And I get the music on hold instead of the ringtone. I have tried
placing Progress() in front of Dial() but to no avail. I do not want
to use the "r" option in Dial() because then I lose the destination
ringtone in early media which is important to my customers.
Anybody had a similar issue? Any idea of what parameters I can try to
tweak, as I am stumped...
Thanks!
Alex
--
_
-- 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
[asterisk-users] security: SIP header spoofing CHANNEL(recvip)?
I am currently suffering various SIP attacks. I am using the following
extension to record the caller's IP address:
exten => h,n,set(CDR(srcip)=${CHANNEL(recvip)})
However, in recent attacks, this IP address is not correct, and I
believe that they are spoofing it. I am using asterisk 1.6.2.15.
Does the CHANNEL(recvip) variable record IP show in the SIP header
instead of the real, UDP source IP? If the CHANNEL(recvip) variable
records the IP address set in the SIP header, and not the real IP
address, how can I obtain the REAL IP address of the caller?
--
_
-- 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
[asterisk-users] Intermitent voice issues
Hi all and thanks for reading. I am experiencing a frustrating issue with asterisk where on some calls the volume suddenly drops to inaudible o completely fades away for a time. If you hold on long enough (20 to 30 seconds) the sound will come back. My asterisk server is on a public IP, and basically acts as a VoIP bridge receiving calls from my customers (all of whom use Grandstream GXW400X gateways on public IP's, no NAT) and sending them to different SIP providers. I am proxying the RTP stream through the server (canreinvite=no). I am sure this is an issue with my setup because this happens to random customers calling to random destinations on any of our VoIP providers. The symptoms are always the same: voice becomes inaudible or fades away, most of the time the customer hangs up. I am running CentOS 5.5 and asterisk 1.6.2.15 on 4 different servers, all of them Dell PowerEdge. Any ideas? I do not even know how to start debugging this issue. A SIP trace obviously shows nothing. Any pointers would be appreciated. Regards, Alex -- _ -- 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
[asterisk-users] DISA does not accept "pause" from cellphones when upgrading from 1.4 to 1.6
I just upgraded my asterisk box from 1.4 + Zaptel to 1.6 + DAHDI and services I was using perfectly before are suddenly broken. I have a DISA access configured, and my companies employees use if to dial into the companies extension from their cell phones. For example they would dial DISA-ACCESS-NUMBER(pause)EXTENSION. Most cellphones (Nokia, Blackberry, iPhone) have some way to save a pause in the dialed string. Nokia does it by saving the letter "p", blackberry has the option to insert a pause (shows up in the address book as "pause"), and the iPhone uses a space and a comma (like DISA-ACCESS-NUMBER, EXTENSION). The pause function on these phones worked perfectly with Asterisk 1.4 + Zaptel. After the upgrade to Asterisk 1.6 + DAHDI, the pause function no longer works. If you MANUALY dial into the access number, wait for the dial tone, and then dial the extension, DISA works perfectly. But if you have the whole DISA access number saved with a pause and an extension and dial, asterisk will not recognize the second number. Basically the DISA dial tone will continue to sound, as if it has not received the extension number. Has anybody else experienced this problem? Any tip would be welcome. Thank you in advance. Alejandro -- _ -- 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] Asterisk not recognizing ACK from an OK message?Help debuging SIP retransmit problem
On Sat, Apr 24, 2010 at 7:01 AM, David White wrote: > > call-id doesn't match? > > SIP/2.0 200 OK > ... > Call-ID: 2117388659-506...@82.158.83.xxx > ... > ACK sip:6615xx...@130.117.xxx.xxx SIP/2.0 > ... > Call-ID: 2117388659-506...@192.168.1.100 > ... > > I'm not sure, but I think that the part after the '@' must also match > throughout the dialog. A Grandstream bug? Thanks! It might be a bug, I'll contact Grandstream and post again if it turns out to be a bug. -- _ -- 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
[asterisk-users] Asterisk not recognizing ACK from an OK message? Help debuging SIP retransmit problem
Hi all. I am having lots of trouble with random calls dropping after 20 seconds, and I finally managed to capture a full sip trace. I'll paste it in full below, but I'll give a summary first. It seems that Asterisk is not recognizing the ACK messages that it receives from the Grandstream ATA. This happens only on the ACK that follows the OK that marks a call as established. This makes Asterisk retransmit the OK message 6 times, after which it drops the call (after exactly 20 seconds). The strange thing is that the "branch" parameter of the ACK is different than the branch parameter from the OK it is replying to, however, this seems to be normal behaviour as specified in the RFC for an ACK that is sent in response to a 200 message. I don't want to clutter the list, and the mail-bot marked as spam my previous email with a pastie link, so I am only including the highlights of the SIP dialog. Any suggestions as to how to attach the whole thing (if necessary) are appreciated. >>> The call was ringing and is now answered: <--- Reliably Transmitting (NAT) to 82.158.83.xxx:5062 ---> SIP/2.0 200 OK Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1226703311;received=82.158.83.xxx;rport=5062 From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@82.158.83.xxx CSeq: 31 INVITE Server: VoIPSwitch Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO Supported: replaces, timer Contact: Content-Type: application/sdp Content-Length: 235 v=0 o=root 1698171141 1698171142 IN IP4 130.117.xxx.xxx s=Asterisk PBX 1.6.1.18 c=IN IP4 130.117.xxx.xxx t=0 0 m=audio 39124 RTP/AVP 18 a=rtpmap:18 G729/8000 a=fmtp:18 annexb=no a=silenceSupp:off - - - - a=ptime:20 a=sendrecv <> -- Packet2Packet bridging SIP/800902-1794 and SIP/130.117.110.21-1795 >>> ATA ACK's the OK message: <--- SIP read from UDP://82.158.83.xxx:5062 ---> ACK sip:6615xx...@130.117.xxx.xxx SIP/2.0 Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1788524356;rport From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@192.168.1.100 CSeq: 31 ACK Contact: Max-Forwards: 70 Supported: replaces, path, timer User-Agent: Grandstream HT-502 V1.1C 1.0.1.57 Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Length: 0 >>> Asterisk does not recognize and retransmits <-> --- (12 headers 0 lines) --- Retransmitting #1 (NAT) to 82.158.83.xxx:5062: SIP/2.0 200 OK Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1226703311;received=82.158.83.xxx;rport=5062 From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@82.158.83.xxx CSeq: 31 INVITE Server: VoIPSwitch Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO Supported: replaces, timer Contact: Content-Type: application/sdp Content-Length: 235 v=0 o=root 1698171141 1698171142 IN IP4 130.117.xxx.xxx s=Asterisk PBX 1.6.1.18 c=IN IP4 130.117.xxx.xxx t=0 0 m=audio 39124 RTP/AVP 18 a=rtpmap:18 G729/8000 a=fmtp:18 annexb=no a=silenceSupp:off - - - - a=ptime:20 a=sendrecv >>> ACK is received again <--- SIP read from UDP://82.158.83.xxx:5062 ---> ACK sip:6615xx...@130.117.xxx.xxx SIP/2.0 Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1788524356;rport From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@192.168.1.100 CSeq: 31 ACK Contact: Max-Forwards: 70 Supported: replaces, path, timer User-Agent: Grandstream HT-502 V1.1C 1.0.1.57 Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Length: 0 ... (cut) >>> The retransmits happen 6 times and then: <-> --- (12 headers 0 lines) --- Retransmitting #6 (NAT) to 82.158.83.xxx:5062: SIP/2.0 200 OK Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1226703311;received=82.158.83.xxx;rport=5062 From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@82.158.83.xxx CSeq: 31 INVITE Server: VoIPSwitch Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO Supported: replaces, timer Contact: Content-Type: application/sdp Content-Length: 235 v=0 o=root 1698171141 1698171142 IN IP4 130.117.xxx.xxx s=Asterisk PBX 1.6.1.18 c=IN IP4 130.117.xxx.xxx t=0 0 m=audio 39124 RTP/AVP 18 a=rtpmap:18 G729/8000 a=fmtp:18 annexb=no a=silenceSupp:off - - - - a=ptime:20 a=sendrecv --- <--- SIP read from UDP://82.158.83.xxx:5062 ---> ACK sip:6615xx...@130.117.xxx.xxx SIP/2.0 Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1788524356;rport From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@192.168.1.100 CSeq: 31 ACK Contact: Max-Forwards: 70 Supported: replaces, path, timer User-Agent: Grandstream HT-502 V1.1C 1.0.1.57 Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Length: 0 <-> --- (12 headers 0 lines) --- <--- SIP read from UDP://82.158.83.xxx:5062 ---> <-> [Apr 23 02:37:15] WARNING[3202]: chan_sip.
[asterisk-users] Asterisk not recognizing ACK from an OK message? Help debuging SIP retransmit problem
Hi all. I am having lots of trouble with random calls dropping after 20 seconds, and I finally managed to capture a full sip trace. I'll paste it in full below, but I'll give a summary first. It seems that Asterisk is not recognizing the ACK messages that it receives from the Grandstream ATA. This happens only on the ACK that follows the OK that marks a call as established. This makes Asterisk retransmit the OK message 6 times, after which it drops the call (after exactly 20 seconds). The strange thing is that the "branch" parameter of the ACK is different than the branch parameter from the OK it is replying to, however, this seems to be normal behaviour as specified in the RFC for an ACK that is sent in response to a 200 message. The full SIP dialog is at http://pastie.org/private/nybdytnfyfenovpwfywcya so as to not clutter the email, but I have included the highlights below: >>> The call was ringing and is now answered: <--- Reliably Transmitting (NAT) to 82.158.83.xxx:5062 ---> SIP/2.0 200 OK Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1226703311;received=82.158.83.xxx;rport=5062 From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@82.158.83.xxx CSeq: 31 INVITE Server: VoIPSwitch Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO Supported: replaces, timer Contact: Content-Type: application/sdp Content-Length: 235 v=0 o=root 1698171141 1698171142 IN IP4 130.117.xxx.xxx s=Asterisk PBX 1.6.1.18 c=IN IP4 130.117.xxx.xxx t=0 0 m=audio 39124 RTP/AVP 18 a=rtpmap:18 G729/8000 a=fmtp:18 annexb=no a=silenceSupp:off - - - - a=ptime:20 a=sendrecv <> -- Packet2Packet bridging SIP/800902-1794 and SIP/130.117.110.21-1795 >>> ATA ACK's the OK message: <--- SIP read from UDP://82.158.83.xxx:5062 ---> ACK sip:6615xx...@130.117.xxx.xxx SIP/2.0 Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1788524356;rport From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@192.168.1.100 CSeq: 31 ACK Contact: Max-Forwards: 70 Supported: replaces, path, timer User-Agent: Grandstream HT-502 V1.1C 1.0.1.57 Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Length: 0 >>> Asterisk does not recognize and retransmits <-> --- (12 headers 0 lines) --- Retransmitting #1 (NAT) to 82.158.83.xxx:5062: SIP/2.0 200 OK Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1226703311;received=82.158.83.xxx;rport=5062 From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@82.158.83.xxx CSeq: 31 INVITE Server: VoIPSwitch Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO Supported: replaces, timer Contact: Content-Type: application/sdp Content-Length: 235 v=0 o=root 1698171141 1698171142 IN IP4 130.117.xxx.xxx s=Asterisk PBX 1.6.1.18 c=IN IP4 130.117.xxx.xxx t=0 0 m=audio 39124 RTP/AVP 18 a=rtpmap:18 G729/8000 a=fmtp:18 annexb=no a=silenceSupp:off - - - - a=ptime:20 a=sendrecv >>> ACK is received again <--- SIP read from UDP://82.158.83.xxx:5062 ---> ACK sip:6615xx...@130.117.xxx.xxx SIP/2.0 Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1788524356;rport From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@192.168.1.100 CSeq: 31 ACK Contact: Max-Forwards: 70 Supported: replaces, path, timer User-Agent: Grandstream HT-502 V1.1C 1.0.1.57 Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Length: 0 ... (cut) >>> The retransmits happen 6 times and then: <-> --- (12 headers 0 lines) --- Retransmitting #6 (NAT) to 82.158.83.xxx:5062: SIP/2.0 200 OK Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1226703311;received=82.158.83.xxx;rport=5062 From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@82.158.83.xxx CSeq: 31 INVITE Server: VoIPSwitch Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO Supported: replaces, timer Contact: Content-Type: application/sdp Content-Length: 235 v=0 o=root 1698171141 1698171142 IN IP4 130.117.xxx.xxx s=Asterisk PBX 1.6.1.18 c=IN IP4 130.117.xxx.xxx t=0 0 m=audio 39124 RTP/AVP 18 a=rtpmap:18 G729/8000 a=fmtp:18 annexb=no a=silenceSupp:off - - - - a=ptime:20 a=sendrecv --- <--- SIP read from UDP://82.158.83.xxx:5062 ---> ACK sip:6615xx...@130.117.xxx.xxx SIP/2.0 Via: SIP/2.0/UDP 82.158.83.xxx:5062;branch=z9hG4bK1788524356;rport From: "800902" ;tag=467506068 To: ;tag=as2e12c791 Call-ID: 2117388659-506...@192.168.1.100 CSeq: 31 ACK Contact: Max-Forwards: 70 Supported: replaces, path, timer User-Agent: Grandstream HT-502 V1.1C 1.0.1.57 Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Length: 0 <-> --- (12 headers 0 lines) --- <--- SIP read from UDP://82.158.83.xxx:5062 ---> <-> [Apr 23 02:37:15] WARNING[3202]: chan_sip.c:3396 retrans_pkt: Maximum retries exceeded on transmission 2117388659-506...@82.158.83.xxx f
Re: [asterisk-users] ${HANGUPCAUSE} is always 0 in the h extension
> However, as I can see by the verbose command, ${HANGUPCAUSE} is always
> 0. I thought it was a channel variable that contained the hangupcause?
Just an update, if the call is established, then there is a
hangupcause received.
The above problem only happens if the caller hangs up before pickup.
This is usualy a cause 16, not 0.
Alex
--
_
-- 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
[asterisk-users] ${HANGUPCAUSE} is always 0 in the h extension
Hi all,
I am using cdr_adaptive_odbc and it works fine. I am trying to save
the q931 hangupcause to a cdr record. My diaplan looks like this.
exten => _X.,1,Dial(${EXTEN})
exten => h,1,Set(CDR(q931)=${HANGUPCAUSE})
exten => h,2,Verbose(${HANGUPCAUSE})
However, as I can see by the verbose command, ${HANGUPCAUSE} is always
0. I thought it was a channel variable that contained the hangupcause?
How can I set this up to correctly save the hangupcause??
Thank you for your help
Regards,
Alex
--
_
-- 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] Calls drop after 20 seconds
> Vieri > Check out the "early dial" feature in the Grandstream products (if you > enabled it) > and play with the "pedantic" option. thanks, already made sure I use pedantic=no and earlydial is off in my GW > Peder > Like the poster below said, do a sip debug on a call and see which end sends > the bye message or ends the call and go from there. That should give you > some sort of clue as to who is having a timer issue. That is my next step, its just so hard to reproduce while debugging! > Stefan > How do you dial the users? direct with the peername or something like > ex...@ipofpeer ? > > i know this problem when dialing a patton ISDN ata without an extension. > The call is established but when the T1 sip timeout fires the call gets > disconnected. Maybe you could do some sip debugging and watch for resend > sip messages. I don't understand, all of my calls are inbound and terminated with different voip carriers, so I am not sure how that will work. I always dial d...@ipofcarrier. Will debug! > Ishfaq > Upgrade phones to latest/most stable firmware > Upgrade routers to latest/most stable firmware This has definetly helped with other problems in the past, so I reccomend it to anybody Thank you so much for all of your help / time guys! -- _ -- 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
[asterisk-users] Time difference in CSV CDR's and MySQL CDR's
Hi all, I am having a curious problem. I use two cdr backends, csv and MySQL. These are my settings: Call Detail Record (CDR) settings -- Logging:Enabled Mode: Batch Log unanswered calls: Yes * Batch Mode Settings --- Safe shutdown: Enabled Threading model:Scheduler plus separate threads Current batch size: 0 records Maximum batch size: 25 records Maximum batch time: 10 seconds Next batch processing time: 7 seconds * Registered Backends --- csv mysql cdr-custom I am finding that the calldate field varies between 3 seconds and 3 minutes between the MySQL database and the CSV files! Is this expected behaviour? I thought they should both use the same timestamp. Is is very difficult to match CDR's this way, and I am finding it hard to trust the results, as I wanted to make sure that my database was behaving correctly and not "losing" any CDR's along the way. Which one of the two CDR's is correct? Should this be posted as a bug? Regards, Alex -- _ -- 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] Unable to load cdr_adaptive_odbc.so
Thanks Tilghman, this immediatley solved the problem. Perhaps a mention in cdr_adaptive_odbc.conf that the res_odbc.so module must also be loaded will help newbies like me ;) -- _ -- 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
[asterisk-users] Unable to load cdr_adaptive_odbc.so
Hi all, I am having trouble getting cdr_adaptive_odbc to work. I have correctly configured the odbc drivers and dsn (I have tested this by connecting directly using isql). I have also configured /etc/asterisk/cdr_adaptive_odbc.conf like so: [test-asterisk] connection=test-asterisk-odbc table=cdr I have tested the ODBC connection test-asterisk-odbc and it works correctly However when I try to load the module I get the following error: asterisk*CLI> module load cdr_adaptive_odbc.so Unable to load module cdr_adaptive_odbc.so Command 'module load cdr_adaptive_odbc.so ' failed. [Apr 21 03:17:30] WARNING[4601]: loader.c:427 load_dynamic_module: Error loading module 'cdr_adaptive_odbc.so': /usr/lib/asterisk/modules/cdr_adaptive_odbc.so: undefined symbol: ast_odbc_request_obj [Apr 21 03:17:30] WARNING[4601]: loader.c:780 load_resource: Module 'cdr_adaptive_odbc.so' could not be loaded. I am stumped, has anybody else run into this problem? Regards, Alex -- _ -- 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] Calls drop after 20 seconds
Doug, thanks for the help, already looked it up, but it does not seem to be a NAT issue (which is what most posters suggest when googling) Danny, those are billsec durations, the call has been established and media is being passed for 20 seconds. Thanks again! Alex -- _ -- 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
[asterisk-users] Calls drop after 20 seconds
Hi all, This issue is giving me a lot of grief with my customers. I have 5 asterisk servers running in production, each one with almost 70 simultaneous calls at peak hour. Most of my customers complain that their calls drop after 20 seconds or so. After running through my cdr's, I see that the number of 20 second calls is MUCH larger than any other number. (see below) billsec count(*) 1 924 2 841 3 725 4 812 5 779 6 681 7 644 8 630 9 613 10 515 11 522 12 516 13 557 14 527 15 507 16 457 17 456 18 467 19 424 20 1644 21 365 22 382 23 353 24 382 25 379 26 370 27 350 28 337 29 302 30 291 >30 lots I am running Asterisk 1.6.1.14 on Debian Lenny. My servers are Dell Power Edge 1950. I use canreinvite=no, and all of my servers are on public IP's. All of my customers use Grandstream GXW4004 telephony adapters. It is a hard issue to debug because it does not happen always. I will try to obtain a sip trace from a dropped call, but until then, any pointers, opinions or even guesses would be much appreciated!! Thanks in advance, Alex -- _ -- 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
[asterisk-users] How to add custom CDR fields to MySQL
Hi all,
I've been trying to add a custom mysql field to my CDR's, but I must
be doing something wrong.
I am using asterisk 1.4 and asterisk 1.6, in extensions.conf I add:
exten => h,1,Set(CDR(q931)=${HANGUPCAUSE})
This extension is executed, I can see it in the asterisk console.
I have added a new column in my MySQL database called q931. However,
the new field does not show up in my database or in the Master.csv
file.
Any help would be greatly appreciated.
Regards,
Alex
--
_
-- 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
[asterisk-users] Unable to forward voice or dtmf
Hi all, I am worried because on my production asterisk servers, I am receiving these errors every 2-3 minutes. my log files are full of them: WARNING[xxx] app_dial.c: Unable to forward voice or dtmf and also, less frequent: WARNING[xxx] app_dial.c: Unable to write frame How can I find out what is causing this problem? If anybody can point me in the right direction I would be very grateful. My server is on a public IP address, and all of my customers are behind NAT. I have set canreinvite=no to keep RTP traffic and thus make it easier to get around NAT issues. All of these calls are terminated at another asterisk server with E1 connections, so there is only one place where NAT is present in our network. Also, all of the calls are inbound (NAT'ed customers call the public IP asterisk server) Regards, Alex -- _ -- 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
[asterisk-users] How to tell if asterisk is handling media or not?
I'm trying to get my asterisk server to reinvite. I have two asterisk servers with public IP's. My users (behind NAT) register on one server (I'll call it server 1), and for some calls they are transfered over to the other server (server 2), because that server has the E1's. I want server 1 to be in the signaling path for billing reasons, but handling the media stream is killing my capacity, and it should not be necessary as server 2 also has a public IP address. I have tried playing around with the "canreinvite" options in sip.conf but the problem is I cannot tell if asterisk is reinviting the call or not. How can I figure out where the media stream is going? thanks! -- _ -- 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] Load balance outgoing calls
Thank you Steve, that's a good idea.
If I use a global variable like
--> IF GLB > 2 GLB = 0
dial(iax2/isp${GLB}/${EXTEN})
--> GLB = GLB +1
I believe this could cause a race condition if two calls are sent to
the carrier at the same time?
--
_
-- 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] Load balance outgoing calls
Hello everybody.
I have a provider that has 3 asterisk boxes which I must balance my
calls against. At the moment, I route different destinations to
different boxes but this causes lots of problems.
Without resorting to OpenSER or other proxies (as my provider also
uses IAX), is there a way I can load balance outgoing channels in
Asterisk?
For example an IAX peer like:
[iax_provider]
type=peer
username=myprovider
host=xxx.xxx.xxx.10
host=xxx.xxx.xxx.11
host=xxx.xxx.xxx.12
secret=verysecret
disalow=all
allow=g729
Is there any way I can balance calls between all of the hosts in the
providers description? In fact, if I set the dialplan like:
exten => _X.,n,Dial(IAX2/iax_provider/${EXTEN}
what IP addres will receive the call? host 10, 11 or 12?
I know DAHDI can balance outgoing calls between the E1's of the span
using DAHDI/r0/ instead of DAHDI/g0. Is there any way of doing this
for other channels?
Thanks!
Alex
--
_
-- 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] agi debug in Asterisk 1.6?
Wow, can't believe I missed that. Thanks so much! -- _ -- 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] agi debug in Asterisk 1.6?
Much to my surprise I tried to debug an AGI script today with "agi debug" on the Asterisk CLI and it did not work. Plus, I could find no reference on lie of it being removed. Is there another name for that command? I scanned the CLI help but found nothing similar. Both my 1.6 boxes do not have the command but my 1.4 box does. Thanks! Alex -- _ -- 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 show peers returns several notices
Hello everybody, When I execute the "sip show peers" command in the asterisk console I always get the following notice, repeated 15 times after the sip show peers output. [Dec 21 03:38:31] NOTICE[12693]: utils.c:1074 ast_wait_for_output: Timed out trying to write This happens on a freshly installed 1.6.1.12 and a 1.6.1.6 box that I am running. Both of them use Debian Linux (lenny) on Dell PowerEdge 1950. My list of SIP peers is quite large (3000+). I have not noticed anything wrong with the asterisk installation apart from this notice, but it is worrying as the error seems to crop up in other bug reports as a precursor to crashes. ___ -- 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] Asterisk 1.6.1.6 crashing
> > *Darrick Hartman:* NO! If you're using a specific 'branch' of asterisk, the latest release > in that branch is the recommended version. There are almost certainly > bugs/issues with earlier versions. 1.6.1.9 is the recommended version > of Asterisk 1.6.1.x. > *Danny Nicholas:* RC's are "bleeding edge"; x.x are considered stable, but you are almost > always better off using the highest x.x stable release. > Ok, I will try upgrading to 1.6.1.9 to see if that helps > *Olivier* No, I didn't mean that : I only meant that I my particular case, that helped > me to work around regular crashes (up to 5 times a day). It doesn't seem our problems are related then, I only have suffered 2 crashes in 1 month of use. Again, thanks for the help guys! ___ -- 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] Asterisk 1.6.1.6 crashing
Hello all, I have a pretty much standard installation of an Asterisk 1.6.1.6 with no PRI cards of any type (full VoIP). Occasionally (it happens every 2 weeks or so), it just stops running. I was using safe_asterisk but it seems that safe_asterisk did not restart it. I do have the core dump file at /tmp/core.myservername-2009-10-20T18:36:20+0200 but it seems it's from an earlier crash. When I read the file, with nano or tail, it seems like pure gibberish. Is there any link or tutorial I can read that will help me read / use core dump files for debugging purposes? This asterisk is under heavy load, it's a 2x Dual core Xeon at 3GHz and its handling around 210 simultaneous calls at peak hour with no transcoding, mostly SIP, but some SIP to IAX conversion, and generaly I feel it's performing well. Do you have any tips on maintenance? Should I restart asterisk every day, or restart the server? Thank you for your help, Alex ___ -- 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
