Re: [OpenSIPS-Users] nat_traversal samples?

[Jeff Chua]

On Thu, Jan 13, 2011 at 3:10 AM, Bogdan-Andrei Iancu
 wrote:
> where is the nat part here ? can you directly route between OpenSIPS B and C
> ?

Bogdan,

I'm trying to use my iPhone from home to route via my PC (vpn). It's
more like this below ...
iPhone, wlan, home PC has private IP. I'm routing from my iPhone to my
PC to the office to get a trunk out.

iPhone - wlan - PC(home) OpenSIPs - VPN - PC(office) OpenSIPS - ISDN

> Regarding TCP, if everything must go through TCP, that will be tricky...you
> can make SIP to go via TCP, but RTP is UDP oriented ...

I know very little about OpenSIPS now, but, again from I read, it
looks like nat_traversal can do the job ... I just need a working
sample if possible. I just can't find it on the net.

Thanks,
Jeff

___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Problem with load balancer module

[Diego Barberio]

Hi Bogdan,

Thanks for your prompt response and all your support. I will correct that
bug, and test again.

Thanks
Diego 

-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
Sent: miércoles, 12 de enero de 2011 05:58 p.m.
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Problem with load balancer module

That's not true - the contact address is the address of the other end 
point, it does not mean that the communication is done directly between 
the end points. The Route hdr is the one dictating the intermediary 
hopsbottom line , contact points the end point, sequential requests 
will visit opensips because of the Route hdr.

Regards,
Bogdan

Diego Barberio wrote:
> Hi Bogdan,
>
> Thanks again for your response.
> I understand your point but as fair as I know if the contact is set to
> 50257609...@192.168.2.165:5061 the subsequent messages (i.e. ACK and BYE)
> will be sent directly to the callee bypassing the proxy. This is OK for
me,
> but I understand that the LB needs to receive the BYE message to keep
track
> of the available resources.
> Please correct me if I'm wrong.
>
> Thanks
> Diego
>
>
> -Original Message-
> From: users-boun...@lists.opensips.org
> [mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
> Sent: miércoles, 12 de enero de 2011 09:15 a.m.
> To: OpenSIPS users mailling list
> Subject: Re: [OpenSIPS-Users] Problem with load balancer module
>
> Hi Diego,
>
> The bug seams to be in your callee device. Take a look at the 200 OK it 
> sends:
>
> U 192.168.2.165:5061 -> 192.168.2.165:5060
> SIP/2.0 200 OK..From: "Your Long 
>
Name";tag=A46E9878A6B36612423768382DD6C758.
> .To: 
>
;tag=843e7f0-a502a8c0-13c5-50022-1110-60a48c5
> 6-1110..Call-ID: 
> d059e3f6f50cacfb33b4526b0a1ca...@192.168.2
> .150..CSeq: 1 INVITE..Via: SIP/2.0/UDP 
> 192.168.2.165;branch=z9hG4bK7bcf.2d3c8236.0..Via: SIP/2.0/UDP 
>
192.168.2.150:5060;received=192.168.2.150;rport=5060;branch=z9hG4bKC115ED423
> 0E704ED2956D13FC3999153..Record-Route: 
>  78A6B36612423768382DD6C758;did=08c.697623b5>..Contact: 
> ..Allow: INVITE, CANCEL, ACK, BYE, 
> OPTIONS, INFO..Content-Type: application/sdp..Content-Length: 
> 210v=0..o=RBTAlerting 2844730 0 IN IP4 192.168.2.165..s=INTEL_
> SIP_CCLIB..i=session information..c=IN IP4 192.168.2.165..t=0 0..m=audio 
> 49152 RTP/AVP 8 101..a=rtpmap:8 PCMA/8000..a=rtpmap:101 
> telephone-event/8000..
> #
>
> The 200 OK is sent from 192.168.2.165:5061, but in contact it places 
> sip:50257609...@192.168.2.165 -> the port is missing, so the sequential 
> requests are going to the wrong destination (to 
> sip:50257609...@192.168.2.165 which is actually the LB ,not the callee)
>
> Regards,
> Bogdan
>
> Diego Barberio wrote:
>   
>> Hi Bogdan,
>>
>> Have you been able to take a look at the traces I sent?
>>
>> Thanks
>> Diego
>>
>> -Original Message-
>> From: users-boun...@lists.opensips.org
>> [mailto:users-boun...@lists.opensips.org] On Behalf Of Diego Barberio
>> Sent: lunes, 10 de enero de 2011 12:51 p.m.
>> To: users@lists.opensips.org
>> Subject: [OpenSIPS-Users] Problem with load balancer module
>>
>> Hi Bogdan,
>>
>> Thank you for your prompt response. I'm sorry I couldn't send the trace
>> before but I had some problems with my network.
>> Also I've change the IP Address schema.
>> The call is originated from 192.168.2.150 to 192.168.2.165:5060 which is
>> 
> the
>   
>> opensips address. Currently, the load balancer is configured to redirect
>> 
> the
>   
>> calls to 192.168.2.165:5061 or 192.168.2.165:5062. In the call I'm
sending
>> the INVITE was redirected to 5061.
>>
>> The you will see that the ACK is not redirected to the destination.
>>
>> Thanks
>> Diego
>>
>> -Original Message-
>> From: users-boun...@lists.opensips.org
>> [mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei
Iancu
>> Sent: viernes, 07 de enero de 2011 01:19 p.m.
>> To: OpenSIPS users mailling list
>> Subject: Re: [OpenSIPS-Users] Problem with load balancer module
>>
>> Hi Diego,
>>
>> Could you post a SIP capture of a complete call (starting with INVITE)
>> 
> from
>   
>> the opensips LB machine ?
>>
>> Regards,
>> Bogdan
>>
>> Diego Barberio wrote:
>>   
>> 
>>> Diego Sebastián Barberio
>>>
>>> www.redmondsoftware.com
>>> +54 11 48153511 (Ext 143)
>>>
>>>
>>> -Original Message-
>>> From: Diego Barberio [mailto:diego.barbe...@redmondsoftware.com]
>>> Sent: jueves, 06 de enero de 2011 03:50 p.m.
>>> To: 'users@lists.opensips.org'
>>> Subject: Problem with load balancer module
>>>
>>> Hello,
>>>
>>> I’m testing the Load Balancing module and I have a problem I can’t fix 
>>> by myself. The INVITE message is routed correctly to one of the
>>> 
>>>   
>> destinations.
>>   
>> 
>>> However the subsequent ACK and the BYE messages are not sent to the 
>>> destinations.
>>>
>>> I set up opensips to run only in the udp 5

Re: [OpenSIPS-Users] Problem with load balancer module

[Bogdan-Andrei Iancu]

That's not true - the contact address is the address of the other end 
point, it does not mean that the communication is done directly between 
the end points. The Route hdr is the one dictating the intermediary 
hopsbottom line , contact points the end point, sequential requests 
will visit opensips because of the Route hdr.


Regards,
Bogdan

Diego Barberio wrote:

Hi Bogdan,

Thanks again for your response.
I understand your point but as fair as I know if the contact is set to
50257609...@192.168.2.165:5061 the subsequent messages (i.e. ACK and BYE)
will be sent directly to the callee bypassing the proxy. This is OK for me,
but I understand that the LB needs to receive the BYE message to keep track
of the available resources.
Please correct me if I'm wrong.

Thanks
Diego


-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
Sent: miércoles, 12 de enero de 2011 09:15 a.m.
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Problem with load balancer module

Hi Diego,

The bug seams to be in your callee device. Take a look at the 200 OK it 
sends:


U 192.168.2.165:5061 -> 192.168.2.165:5060
SIP/2.0 200 OK..From: "Your Long 
Name";tag=A46E9878A6B36612423768382DD6C758.
.To: 
;tag=843e7f0-a502a8c0-13c5-50022-1110-60a48c5
6-1110..Call-ID: 
d059e3f6f50cacfb33b4526b0a1ca...@192.168.2
.150..CSeq: 1 INVITE..Via: SIP/2.0/UDP 
192.168.2.165;branch=z9hG4bK7bcf.2d3c8236.0..Via: SIP/2.0/UDP 
192.168.2.150:5060;received=192.168.2.150;rport=5060;branch=z9hG4bKC115ED423
0E704ED2956D13FC3999153..Record-Route: 
78A6B36612423768382DD6C758;did=08c.697623b5>..Contact: 
..Allow: INVITE, CANCEL, ACK, BYE, 
OPTIONS, INFO..Content-Type: application/sdp..Content-Length: 
210v=0..o=RBTAlerting 2844730 0 IN IP4 192.168.2.165..s=INTEL_
SIP_CCLIB..i=session information..c=IN IP4 192.168.2.165..t=0 0..m=audio 
49152 RTP/AVP 8 101..a=rtpmap:8 PCMA/8000..a=rtpmap:101 
telephone-event/8000..

#

The 200 OK is sent from 192.168.2.165:5061, but in contact it places 
sip:50257609...@192.168.2.165 -> the port is missing, so the sequential 
requests are going to the wrong destination (to 
sip:50257609...@192.168.2.165 which is actually the LB ,not the callee)


Regards,
Bogdan

Diego Barberio wrote:
  

Hi Bogdan,

Have you been able to take a look at the traces I sent?

Thanks
Diego

-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Diego Barberio
Sent: lunes, 10 de enero de 2011 12:51 p.m.
To: users@lists.opensips.org
Subject: [OpenSIPS-Users] Problem with load balancer module

Hi Bogdan,

Thank you for your prompt response. I'm sorry I couldn't send the trace
before but I had some problems with my network.
Also I've change the IP Address schema.
The call is originated from 192.168.2.150 to 192.168.2.165:5060 which is


the
  

opensips address. Currently, the load balancer is configured to redirect


the
  

calls to 192.168.2.165:5061 or 192.168.2.165:5062. In the call I'm sending
the INVITE was redirected to 5061.

The you will see that the ACK is not redirected to the destination.

Thanks
Diego

-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
Sent: viernes, 07 de enero de 2011 01:19 p.m.
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Problem with load balancer module

Hi Diego,

Could you post a SIP capture of a complete call (starting with INVITE)


from
  

the opensips LB machine ?

Regards,
Bogdan

Diego Barberio wrote:
  


Diego Sebastián Barberio

www.redmondsoftware.com
+54 11 48153511 (Ext 143)


-Original Message-
From: Diego Barberio [mailto:diego.barbe...@redmondsoftware.com]
Sent: jueves, 06 de enero de 2011 03:50 p.m.
To: 'users@lists.opensips.org'
Subject: Problem with load balancer module

Hello,

I’m testing the Load Balancing module and I have a problem I can’t fix 
by myself. The INVITE message is routed correctly to one of the

  

destinations.
  

However the subsequent ACK and the BYE messages are not sent to the 
destinations.


I set up opensips to run only in the udp 5060 port. Then I have two 
identical applications: one running on port 5061 and the other on port 
5062, the 3 components are running in the same server which has a 
single IP

address: 192.168.1.195.
The application is very simple:
1. Receives the INVITE, starts streaming the RTP, and sends the OK
2. When the ACK is received injects some music in the streaming
3. Waits until de BYE is received. Then stops the streaming and

  
sends 
  


the OK.

This is the configuration of the load_balancer table:

mysql> select * from load_balancer;
++--++---++---
++--++---++---
++--+

Re: [OpenSIPS-Users] Problem with load balancer module

[Diego Barberio]

Hi Bogdan,

Thanks again for your response.
I understand your point but as fair as I know if the contact is set to
50257609...@192.168.2.165:5061 the subsequent messages (i.e. ACK and BYE)
will be sent directly to the callee bypassing the proxy. This is OK for me,
but I understand that the LB needs to receive the BYE message to keep track
of the available resources.
Please correct me if I'm wrong.

Thanks
Diego


-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
Sent: miércoles, 12 de enero de 2011 09:15 a.m.
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Problem with load balancer module

Hi Diego,

The bug seams to be in your callee device. Take a look at the 200 OK it 
sends:

U 192.168.2.165:5061 -> 192.168.2.165:5060
SIP/2.0 200 OK..From: "Your Long 
Name";tag=A46E9878A6B36612423768382DD6C758.
.To: 
;tag=843e7f0-a502a8c0-13c5-50022-1110-60a48c5
6-1110..Call-ID: 
d059e3f6f50cacfb33b4526b0a1ca...@192.168.2
.150..CSeq: 1 INVITE..Via: SIP/2.0/UDP 
192.168.2.165;branch=z9hG4bK7bcf.2d3c8236.0..Via: SIP/2.0/UDP 
192.168.2.150:5060;received=192.168.2.150;rport=5060;branch=z9hG4bKC115ED423
0E704ED2956D13FC3999153..Record-Route: 
..Contact: 
..Allow: INVITE, CANCEL, ACK, BYE, 
OPTIONS, INFO..Content-Type: application/sdp..Content-Length: 
210v=0..o=RBTAlerting 2844730 0 IN IP4 192.168.2.165..s=INTEL_
SIP_CCLIB..i=session information..c=IN IP4 192.168.2.165..t=0 0..m=audio 
49152 RTP/AVP 8 101..a=rtpmap:8 PCMA/8000..a=rtpmap:101 
telephone-event/8000..
#

The 200 OK is sent from 192.168.2.165:5061, but in contact it places 
sip:50257609...@192.168.2.165 -> the port is missing, so the sequential 
requests are going to the wrong destination (to 
sip:50257609...@192.168.2.165 which is actually the LB ,not the callee)

Regards,
Bogdan

Diego Barberio wrote:
> Hi Bogdan,
>
> Have you been able to take a look at the traces I sent?
>
> Thanks
> Diego
>
> -Original Message-
> From: users-boun...@lists.opensips.org
> [mailto:users-boun...@lists.opensips.org] On Behalf Of Diego Barberio
> Sent: lunes, 10 de enero de 2011 12:51 p.m.
> To: users@lists.opensips.org
> Subject: [OpenSIPS-Users] Problem with load balancer module
>
> Hi Bogdan,
>
> Thank you for your prompt response. I'm sorry I couldn't send the trace
> before but I had some problems with my network.
> Also I've change the IP Address schema.
> The call is originated from 192.168.2.150 to 192.168.2.165:5060 which is
the
> opensips address. Currently, the load balancer is configured to redirect
the
> calls to 192.168.2.165:5061 or 192.168.2.165:5062. In the call I'm sending
> the INVITE was redirected to 5061.
>
> The you will see that the ACK is not redirected to the destination.
>
> Thanks
> Diego
>
> -Original Message-
> From: users-boun...@lists.opensips.org
> [mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
> Sent: viernes, 07 de enero de 2011 01:19 p.m.
> To: OpenSIPS users mailling list
> Subject: Re: [OpenSIPS-Users] Problem with load balancer module
>
> Hi Diego,
>
> Could you post a SIP capture of a complete call (starting with INVITE)
from
> the opensips LB machine ?
>
> Regards,
> Bogdan
>
> Diego Barberio wrote:
>   
>> Diego Sebastián Barberio
>>
>> www.redmondsoftware.com
>> +54 11 48153511 (Ext 143)
>>
>>
>> -Original Message-
>> From: Diego Barberio [mailto:diego.barbe...@redmondsoftware.com]
>> Sent: jueves, 06 de enero de 2011 03:50 p.m.
>> To: 'users@lists.opensips.org'
>> Subject: Problem with load balancer module
>>
>> Hello,
>>
>> I’m testing the Load Balancing module and I have a problem I can’t fix 
>> by myself. The INVITE message is routed correctly to one of the
>> 
> destinations.
>   
>> However the subsequent ACK and the BYE messages are not sent to the 
>> destinations.
>>
>> I set up opensips to run only in the udp 5060 port. Then I have two 
>> identical applications: one running on port 5061 and the other on port 
>> 5062, the 3 components are running in the same server which has a 
>> single IP
>> address: 192.168.1.195.
>> The application is very simple:
>>  1. Receives the INVITE, starts streaming the RTP, and sends the OK
>>  2. When the ACK is received injects some music in the streaming
>>  3. Waits until de BYE is received. Then stops the streaming and
>> 
> sends 
>   
>> the OK.
>>
>> This is the configuration of the load_balancer table:
>>
>> mysql> select * from load_balancer;
>> ++--++---++---
>> ++--++---++---
>> ++--++---++---
>> +
>> | id | group_id | dst_uri| resources | probe_mode |
>> description |
>> ++--++---++---
>> ++--++---++---
>> +-

Re: [OpenSIPS-Users] nat_traversal samples?

[Bogdan-Andrei Iancu]

Jeff Chua wrote:

On Wed, Jan 12, 2011 at 7:56 PM, Bogdan-Andrei Iancu
 wrote:
  

Hi Jeff,

The UDP versus TCP issue is for the communication between opensips and media
relay ? If so, how comes you have a firewall between them ? you have
opensips and the media relay in different networks ?



Bogdan,

I'm looking for something like this ...

SIP Client A -- OpenSIPS B -- VPN (only TCP/IP allowed) -- OpenSIPS C
-- SIP Client D

I read that with nat_travesal, I won't even need rtpproxy or mediaproxy.
  

Jeff,

where is the nat part here ? can you directly route between OpenSIPS B 
and C ?


Regarding TCP, if everything must go through TCP, that will be 
tricky...you can make SIP to go via TCP, but RTP is UDP oriented ...


Regards,
Bogdan



Thanks,
Jeff

___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

  



--
Bogdan-Andrei Iancu
OpenSIPS Event - expo, conf, social, bootcamp
2 - 4 February 2011, ITExpo, Miami,  USA
www.voice-system.ro


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Dialog Module and Bogus Event 8 in state 2

[Bogdan-Andrei Iancu]

Hi Sven,

Interesting, never saw a NOTIFY in early state of a dialog...can you 
post a SIP capture for such a dialog ?


Going back to your question, the message said that the notify event does 
not fit to the current dialog state, but this has no effect on the 
dialog state, neither on the routing / processing of the NOTIFY...


Best regards,
Bogdan

Sven Schulz wrote:

Directly after the 183 Ringing is a NOTIFY message coming from the
destination (which is a Cisco Sip gateway). The source sends a corresponding
200 OK to this NOTIFY (also a cisco PBX).

So is the "bogus event" something I should be concerned with or is it more
of an informational error message?

0.0010.1.2.52 -> 10.1.1.82SIP/SDP Request: INVITE
sip:phonenum...@xx.edu:5060, with session description
  0.00029110.1.1.82 -> 10.1.2.52SIP Status: 100 Giving a try
  0.00183010.1.1.82 -> 10.1.1.60SIP/SDP Request: INVITE
sip:18148656...@10.1.1.60, with session description
  0.00434310.1.1.60 -> 10.1.1.82SIP Status: 100 Trying
  1.47410010.1.1.60 -> 10.1.1.82SIP/SDP Status: 183 Session
Progress, with session description
  1.47424010.1.1.82 -> 10.1.2.52SIP/SDP Status: 183 Session
Progress, with session description
  1.48385010.1.1.60 -> 10.1.1.82SIP Request: NOTIFY
sip:10.1.2.52:5060
  1.48396210.1.1.82 -> 10.1.2.52SIP Request: NOTIFY
sip:10.1.2.52:5060
  1.48609110.1.2.52 -> 10.1.1.82SIP Status: 200 OK
  1.48613310.1.1.82 -> 10.1.1.60SIP Status: 200 OK
  4.25237310.1.1.60 -> 10.1.1.82SIP/SDP Status: 200 OK, with session
description
  4.25275610.1.1.82 -> 10.1.2.52SIP/SDP Status: 200 OK, with session
description



On 1/12/11 6:19 AM, "Bogdan-Andrei Iancu"  wrote:

  

Hi Sven,

"Bogus Event 8 in state 2" is translated to receiving an indialog
request (non ACK, non BYE) while dialog in early state.maybe it is a
PRACK to the 183can you check that ?

Regards,
Bogdan

Sven Schulz wrote:


Running opensips 1.6.3, dialog module seems to function correctly
however I keep getting these messages:

CRITICAL:dialog:log_next_state_dlg: bogus event 8 in state 2 for dlg
0x2b33956052c0 [2284:61359203] with clid
'11a0bd80-d2c160b0-1a-34020...@10.1.2.52
<%2711a0bd80-d2c160b0-1a-34020...@10.1.2.52>' and tags
'd5edda48-9c10-424c-b200-8ec1eb8e532c-42504961' '292F5834-D13'

They only seem to happen when an INVITE is followed by a 183 RINGING
message. INVITES without a 183 wont get this error messege. Is this
normal or should I be concerned?


Sven Schulz
Penn State University
Telecommunications and Network Services
814.865.6116
sip:s...@psu.edu



___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
  
  



___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

  



--
Bogdan-Andrei Iancu
OpenSIPS Event - expo, conf, social, bootcamp
2 - 4 February 2011, ITExpo, Miami,  USA
www.voice-system.ro


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Example config for NATed UACs, RTPproxy, and NATed OpenSIPS (version 1.6.4)

[Bogdan-Andrei Iancu]

James, never user openVZ so far..there are a log of VM technologies out 
there :)For the moment we release the opensips live distro on VMware 
as that;s the main what we used...not sure what are the other main VM 
tech used by other people...


Regards,
Bogdan

James Lamanna wrote:

Bogdan,
Wow, I didn't know about the live DVD.
Any chance someone could create this as an OpenVZ container in
addition to VMWare?

-- James

On Mon, Jan 10, 2011 at 2:25 AM, Bogdan-Andrei Iancu
 wrote:
  

Hi Damon,

Well, the answer is simple - download the opensips virtual machine
(http://www.voice-system.ro/shortcuts::opensips_livedvd)  were you have a
ready to run opensips platform with NAT traversal support - you can see in
the script form the VM how the NAT traversal is done (for signalling and
media).

If you have questions on that, please come back here.

Regards,
Bogdan

Damon Miller wrote:


All,


I've seen many requests for an example working config that shows a working
RTPproxy configuration with NATed clients, but I haven't seen many
responses.  I recently spent an absurd amount of time getting a working
configuration in place so I thought I would post it here in case it's
helpful to anyone.

Three quick points:

1.  I have only tested this with clients behind a NAT firewall, i.e. I
haven't tested with clients that have a public IP.


2.  My OpenSIPS server is behind a NAT firewall itself.  To deal with
this, I added the two "advertised" options, as follows:

advertised_address="xx.xx.xx.xx"
alias="xx.xx.xx.xx:5060


(Replace the "xx.xx.xx.xx" with the NAT firewall's public IP.)

I also had to use a modified version of RTPproxy that presents the
firewall's public IP even though it binds to a private IP.  Here's a post
which summarizes that version of RTPproxy:


http://opensips-open-sip-server.1449251.n2.nabble.com/Rtpproxy-behind-the-NAT-td5008041.html


I run RTPproxy like this:

rtpproxy -A xx.xx.xx.xx -l 192.168.20.154 -s udp:127.0.0.1:12221 -m 25000
-M 65000 -F -d DBUG:LOCAL1


3.  I had to "tell" OpenSIPS that my firewall's public IP was one of its
local domains.  I'm using MySQL as you'll see in the config file so all I
had to do was insert a value into the 'domain' table.  That was pretty
obvious, i.e.:

mysql> insert into domain (domain) values ("xx.xx.xx.xx");

(Replace 'xx.xx.xx.xx' with your public IP.)



Here's my 'opensips.cfg' file:

--

# --- global configuration parameters 
debug=3
fork=yes
log_facility=LOG_LOCAL0
log_stderror=no
children=4
port=5060
dns=no
rev_dns=no

advertised_address="xx.xx.xx.xx"
alias="xx.xx.xx.xx:5060"

# -- module loading --
mpath="/usr/local/lib64/opensips/modules/"
loadmodule "db_mysql.so"
loadmodule "signaling.so"
loadmodule "sl.so"
loadmodule "tm.so"
loadmodule "rr.so"
loadmodule "maxfwd.so"
loadmodule "usrloc.so"
loadmodule "registrar.so"
loadmodule "textops.so"
loadmodule "mi_fifo.so"
loadmodule "uri.so"
loadmodule "nathelper.so"
loadmodule "domain.so"

# - setting module-specific parameters ---
modparam("mi_fifo", "fifo_name", "/tmp/opensips_fifo")
modparam("usrloc", "db_url",
"mysql://opensipsrw:opensip...@localhost/opensips")
modparam("usrloc", "db_mode", 2)
modparam("rr", "enable_full_lr", 1)
modparam("nathelper", "rtpproxy_sock", "udp:127.0.0.1:12221")
modparam("nathelper", "nortpproxy_str", "")
modparam("domain", "db_url",
"mysql://opensipsrw:opensip...@localhost/opensips")

## NAT ##
modparam("usrloc", "nat_bflag", 6)
modparam("nathelper", "ping_nated_only", 1)
modparam("nathelper", "sipping_bflag", 8)
modparam("nathelper", "received_avp", "$avp(i:801)")
## NAT ##


# main routing logic
route {

   # initial sanity checks
   if (!mf_process_maxfwd_header("10")) {
   sl_send_reply("483","Too Many Hops");
   exit;
   };

   if (msg:len >=  2048 ) {
   sl_send_reply("513", "Message too big");
   exit;
   };


   ## NAT ##
   if (nat_uac_test("3")) {

   if (is_method("REGISTER") && !is_present_hf("Record-Route")) {

   # Rewrite contact with source IP of signalling
   fix_nated_contact();

   force_rport();
   setbflag(6); # Mark as NATed

   # if you want SIP NAT pinging
   setbflag(8);
   };
   };
   ## NAT ##

   if (!method=="REGISTER")
   record_route();

   # subsequent messages withing a dialog should take the
   # path determined by record-routing
   if (loose_route()) {
   # mark routing logic in request
   append_hf("P-hint: rr-enforced\r\n");
   route(1);
   };

   if (!uri==myself) {
   # mark routing logic in request
   append_hf("P-hint: outbound\r\n");
   route(1);
   };

   if (uri==myself) {
   if (method=="REGISTER") {
   save("location");
   exit;
   };
   }

   if (

Re: [OpenSIPS-Users] Example config for NATed UACs, RTPproxy, and NATed OpenSIPS (version 1.6.4)

[James Lamanna]

Bogdan,
Wow, I didn't know about the live DVD.
Any chance someone could create this as an OpenVZ container in
addition to VMWare?

-- James

On Mon, Jan 10, 2011 at 2:25 AM, Bogdan-Andrei Iancu
 wrote:
> Hi Damon,
>
> Well, the answer is simple - download the opensips virtual machine
> (http://www.voice-system.ro/shortcuts::opensips_livedvd)  were you have a
> ready to run opensips platform with NAT traversal support - you can see in
> the script form the VM how the NAT traversal is done (for signalling and
> media).
>
> If you have questions on that, please come back here.
>
> Regards,
> Bogdan
>
> Damon Miller wrote:
>>
>> All,
>>
>>
>> I've seen many requests for an example working config that shows a working
>> RTPproxy configuration with NATed clients, but I haven't seen many
>> responses.  I recently spent an absurd amount of time getting a working
>> configuration in place so I thought I would post it here in case it's
>> helpful to anyone.
>>
>> Three quick points:
>>
>> 1.  I have only tested this with clients behind a NAT firewall, i.e. I
>> haven't tested with clients that have a public IP.
>>
>>
>> 2.  My OpenSIPS server is behind a NAT firewall itself.  To deal with
>> this, I added the two "advertised" options, as follows:
>>
>> advertised_address="xx.xx.xx.xx"
>> alias="xx.xx.xx.xx:5060
>>
>>
>> (Replace the "xx.xx.xx.xx" with the NAT firewall's public IP.)
>>
>> I also had to use a modified version of RTPproxy that presents the
>> firewall's public IP even though it binds to a private IP.  Here's a post
>> which summarizes that version of RTPproxy:
>>
>>
>> http://opensips-open-sip-server.1449251.n2.nabble.com/Rtpproxy-behind-the-NAT-td5008041.html
>>
>>
>> I run RTPproxy like this:
>>
>> rtpproxy -A xx.xx.xx.xx -l 192.168.20.154 -s udp:127.0.0.1:12221 -m 25000
>> -M 65000 -F -d DBUG:LOCAL1
>>
>>
>> 3.  I had to "tell" OpenSIPS that my firewall's public IP was one of its
>> local domains.  I'm using MySQL as you'll see in the config file so all I
>> had to do was insert a value into the 'domain' table.  That was pretty
>> obvious, i.e.:
>>
>> mysql> insert into domain (domain) values ("xx.xx.xx.xx");
>>
>> (Replace 'xx.xx.xx.xx' with your public IP.)
>>
>>
>>
>> Here's my 'opensips.cfg' file:
>>
>> --
>>
>> # --- global configuration parameters 
>> debug=3
>> fork=yes
>> log_facility=LOG_LOCAL0
>> log_stderror=no
>> children=4
>> port=5060
>> dns=no
>> rev_dns=no
>>
>> advertised_address="xx.xx.xx.xx"
>> alias="xx.xx.xx.xx:5060"
>>
>> # -- module loading --
>> mpath="/usr/local/lib64/opensips/modules/"
>> loadmodule "db_mysql.so"
>> loadmodule "signaling.so"
>> loadmodule "sl.so"
>> loadmodule "tm.so"
>> loadmodule "rr.so"
>> loadmodule "maxfwd.so"
>> loadmodule "usrloc.so"
>> loadmodule "registrar.so"
>> loadmodule "textops.so"
>> loadmodule "mi_fifo.so"
>> loadmodule "uri.so"
>> loadmodule "nathelper.so"
>> loadmodule "domain.so"
>>
>> # - setting module-specific parameters ---
>> modparam("mi_fifo", "fifo_name", "/tmp/opensips_fifo")
>> modparam("usrloc", "db_url",
>> "mysql://opensipsrw:opensip...@localhost/opensips")
>> modparam("usrloc", "db_mode", 2)
>> modparam("rr", "enable_full_lr", 1)
>> modparam("nathelper", "rtpproxy_sock", "udp:127.0.0.1:12221")
>> modparam("nathelper", "nortpproxy_str", "")
>> modparam("domain", "db_url",
>> "mysql://opensipsrw:opensip...@localhost/opensips")
>>
>> ## NAT ##
>> modparam("usrloc", "nat_bflag", 6)
>> modparam("nathelper", "ping_nated_only", 1)
>> modparam("nathelper", "sipping_bflag", 8)
>> modparam("nathelper", "received_avp", "$avp(i:801)")
>> ## NAT ##
>>
>>
>> # main routing logic
>> route {
>>
>>    # initial sanity checks
>>    if (!mf_process_maxfwd_header("10")) {
>>        sl_send_reply("483","Too Many Hops");
>>        exit;
>>    };
>>
>>    if (msg:len >=  2048 ) {
>>        sl_send_reply("513", "Message too big");
>>        exit;
>>    };
>>
>>
>>    ## NAT ##
>>    if (nat_uac_test("3")) {
>>
>>        if (is_method("REGISTER") && !is_present_hf("Record-Route")) {
>>
>>            # Rewrite contact with source IP of signalling
>>            fix_nated_contact();
>>
>>            force_rport();
>>            setbflag(6); # Mark as NATed
>>
>>            # if you want SIP NAT pinging
>>            setbflag(8);
>>        };
>>    };
>>    ## NAT ##
>>
>>    if (!method=="REGISTER")
>>        record_route();
>>
>>    # subsequent messages withing a dialog should take the
>>    # path determined by record-routing
>>    if (loose_route()) {
>>        # mark routing logic in request
>>        append_hf("P-hint: rr-enforced\r\n");
>>        route(1);
>>    };
>>
>>    if (!uri==myself) {
>>        # mark routing logic in request
>>        append_hf("P-hint: outbound\r\n");
>>        route(1);

Re: [OpenSIPS-Users] new install and INVITES not being forwarded

[Gareth Blades]

I installed a caching nameserver but it made no difference.
I then switched logging from syslog to stderr and thats much better and 
the phone can register fine.
I have attached the debug from when I tried making the call between 
lines. If there is nothing usefull there then can you let me know where 
I could put some additional loggin in the config file in order to help.


Thanks

Bogdan-Andrei Iancu wrote:

Hi Gareth,

Gareth Blades wrote:
I am having a problem with running opensips in debug level 6. When 
opensips is set to this I am finding that it takes a long time to 
respond to register requests (over 5 seconds compared to a fraction of 
a second) which means that my phone times out when trying to register 
so I cannot then debug trying to send calls through.
woow...that is really strange.never encountered something like 
thatdo you log to stderror(console) or to syslog ?


Any ideas?

DNS timeouts are a common cause for these sort of pauses. Are there 
any specific DNS setup requirements that opensips has that I might 
have missing?

see the DNS related params:
   http://www.opensips.org/Resources/DocsCoreFcn#toc41

Also, better use DNS cache app between your opensips and the outer world.


So, your missing INVITE (not being sent out) are because of opensips 
blocking in some DNS query ?


Regards,
Bogdan


Thanks
Gareth

Bogdan-Andrei Iancu wrote:

Hi Gareth,

On a first look, the script looks ok, but as a general way to 
debug/troubleshoot your script, place xlog()'s in your script, in 
different points, to see if the script execution gets to that point.


Regarding the 408 - take care your script does not have NAT support, 
so it may not work if your client is behind a NAT.


Regards,
Bogdan

Gareth Blades wrote:

Thanks I will need to get back to you on Monday.

The phone on my desk has just started not being able to register. 
Its showing registration status 408 which is no response.
I can see the opensips server receiving the the registration and 
issuing a 401 unauthorised but I dont think this is getting back 
through the firewall.
I will get some static forwarding added to the firewall to avoid 
this sort of issue complicaing investigating the other problem.


In the meantime I have attached my config file.


Bogdan-Andrei Iancu wrote:

Hi Gareth,

looking at the logs, it seams that your script processing (for the 
INVITE) never get to a t_relay() point - place some xlog() prints 
in your script to see where the INVITE processing is going through.


Regards,
Bogdan

Gareth Blades wrote:
Thanks. I have attached a text file as its a bit long to paste in 
a message and there are long lines.


Looking through the log these couple of entries look significant.

DBG:auth:build_auth_hf: 'Proxy-Authenticate: Digest 
realm="vmopensips1.skycomuk.com", 
nonce="4d26db660001d69ce33126031746bdf4735e3f0f922d"

...
DBG:tm:matching_3261: RFC3261 transaction matching failed




Bogdan-Andrei Iancu wrote:

Hi Gareth,

hard to say without some logs - do you see errors in the opensips 
log ? if not, increase the debug level to 6 and post the logs 
corresponding the call.


Regards,
Bogdan

Gareth Blades wrote:

Setup:-
Server installed at a hosting facility with its owm public IP 
address.
A desk phone in the office with a couple of accounts configured 
registered to opensips through a NAT firewall.
The opensips server has been installed and configured using the 
guide in the Opensips PACT book together with database support 
(no multidomain).


The two lines on the phone are able to register fine and I can 
see them as being registered in the database and 'opensipsctrl 
ul show' displays what I would expect.


However when I call between the two lines and minitoring via 
ngrep I see the initial INVITE go out and then a '407 proxy 
authentication required' comes back and the phone sends the 
invite again with the authentication but opensips doesnt send 
any reply. The phone retransmits the packet a couple of times 
and displays call failed.


Any idea what may be wrong?


I know the call audio is unlikely to work but that will be the 
next step. I am just trying to get each step working as I go at 
the moment.



Thanks
Gareth

___






___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users






Jan 12 16:03:17 [20309] DBG:core:parse_msg: SIP Request:
Jan 12 16:03:17 [20309] DBG:core:parse_msg:  method:  
Jan 12 16:03:17 [20309] DBG:core:parse_msg:  uri: 

Jan 12 16:03:17 [20309] DBG:core:parse_msg:  version: 
Jan 12 16:03:17 [20309] DBG:core:parse_headers: flags=2
Jan 12 16:03:17 [20309] DBG:core:parse_via_param: found param type 232, 
 = ; state=16
Jan 12 16:03:17 [20309] DBG:core:parse_via: end of header reached, state=5
Jan 12 16:03:17 [20309] DBG:core:parse_headers: via found, flags=2
Jan 12 16:03:17 [20309] DBG:core:parse_headers: this is the first v

Re: [OpenSIPS-Users] Pacth rtpproxy

[Razvan Crainea]

Hello Denis,

 A call is established when the callee answers it.

Regards,
Razvan

On 01/12/2011 04:52 PM, Denis Putyato wrote:


Hello, Razvan

“This is a problem since we would like to have a long period for call 
establishment” and what does it mean “call establishment” in such context?


*From:*users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] *On Behalf Of *Razvan Crainea

*Sent:* Wednesday, January 12, 2011 5:18 PM
*To:* OpenSIPS users mailling list
*Subject:* Re: [OpenSIPS-Users] Pacth rtpproxy

Hello Denis,

RTPProxy is only used to detect the media timeout. If OpenSIPS 
receives a timeout notification on an unestablished call, it simply 
ignores it.
If you want to terminate the call when the callee doesn't answer you 
can use the tm module and set the "fr_inv_timer" parameter. You can 
get more details from:

http://www.opensips.org/html/docs/modules/devel/tm.html#id250344

Regards,
Razvan

On 01/12/2011 02:38 PM, Denis Putyato wrote:

Razvan, I got rtpproxy from http://opensips.org/pub/rtpproxy/ as you 
wrote.


I started it using such cli command “/usr/local/rtpproxy1/bin/rtpproxy 
-u opensips -l 1.1.1.1 -s /var/run/rtpproxy.sock -T 80 -i -n 
/var/run/timer.sock  -d INFO” and made test call.


Callee has been ringing during about 2 minutes and nothing happens at 
all. What I did wrong?


P.S. I use such function in my script for rtp proxy 
“rtpproxy_offer("con");”


*From:*users-boun...@lists.opensips.org 
 
[mailto:users-boun...@lists.opensips.org] *On Behalf Of *Razvan Crainea

*Sent:* Wednesday, January 12, 2011 3:14 PM
*To:* OpenSIPS users mailling list
*Subject:* Re: [OpenSIPS-Users] Pacth rtpproxy

Hello Denis,

In the official release of RTPProxy, the timeout parameter (-T) 
controls both session establishment and rtp timeout. This is a problem 
since we would like to have a long period for call establishment, but 
a fast media timeout detection.
In the patched version of RTPProxy, the -W parameter allows you to 
specify a longer period for call establishment. If not set, it has the 
default value of 60 seconds.
If you decide not to use patched version of RTPProxy, the timeout 
notification will work, but you will have the same timeout in both 
situations.


Regards,
Razvan


On 01/12/2011 07:38 AM, Denis Putyato wrote:

Hello Razvan,

“OpenSIPS shouldn't even try to terminate the call because it isn't 
established yet”


As I understand I just do not need to use –W key when starting 
rtpproxy, it does not work at all?


*From:*users-boun...@lists.opensips.org 
 
[mailto:users-boun...@lists.opensips.org] *On Behalf Of *Razvan Crainea

*Sent:* Tuesday, January 11, 2011 6:49 PM
*To:* OpenSIPS users mailling list
*Subject:* Re: [OpenSIPS-Users] Pacth rtpproxy

Hello Denis,

You are right, OpenSIPS shouldn't even try to terminate the call 
because it isn't established yet. I just added a small fix to solve 
this problem. Please update your code from svn to use this fix.
The RTPProxy patch was done against commit 
"600c80493793bafd2d69427bc22fcb43faad98c5". You can either get the 
RTPProxy from git, change it's branch and then apply the patch, or you 
can download an already patched version from 
http://opensips.org/pub/rtpproxy/.


Regards,
Razvan

On 1/11/2011 2:19 PM, Denis Putyato wrote:

Hello!

I try patch rtpproxy gotten from git. And there is such error during 
patching


patch < rtpproxy_timeout_notification.patch

patching file main.c

Hunk #1 succeeded at 70 (offset 2 lines).

Hunk #2 FAILED at 120.

Hunk #3 succeeded at 132 with fuzz 1 (offset 4 lines).

Hunk #4 succeeded at 211 with fuzz 2 (offset 4 lines).

Hunk #5 succeeded at 276 (offset 4 lines).

Hunk #6 succeeded at 742 with fuzz 2 (offset -26 lines).

Hunk #7 succeeded at 758 with fuzz 2 (offset -26 lines).

1 out of 7 hunks FAILED -- saving rejects to file main.c.rej

patching file rtpp_command.c

Hunk #1 FAILED at 795.

Hunk #2 FAILED at 888.

2 out of 2 hunks FAILED -- saving rejects to file rtpp_command.c.rej

patching file rtpp_defines.h

Hunk #1 FAILED at 95.

1 out of 1 hunk FAILED -- saving rejects to file rtpp_defines.h.rej

patching file rtpp_notify.c

rtpproxy_timeout_notification.patch is a patch for timeout 
notification which  divide rtp timeout and session initiation timeout 
notification as said in


http://www.opensips.org/html/docs/modules/devel/nathelper.html#id249142

This patch I got from SVN version of latest Opensips.

  
  
___

Users mailing list
Users@lists.opensips.org  
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

  
  
___

Users mailing list
Users@lists.opensips.org  
http://lists.opensips.org/cgi-bin/mailman/listinfo/users




--
Razvan Crainea
www.voice-system.ro  
  
  
___

Re: [OpenSIPS-Users] Dialog Module and Bogus Event 8 in state 2

[Sven Schulz]

Directly after the 183 Ringing is a NOTIFY message coming from the
destination (which is a Cisco Sip gateway). The source sends a corresponding
200 OK to this NOTIFY (also a cisco PBX).

So is the "bogus event" something I should be concerned with or is it more
of an informational error message?

0.0010.1.2.52 -> 10.1.1.82SIP/SDP Request: INVITE
sip:phonenum...@xx.edu:5060, with session description
  0.00029110.1.1.82 -> 10.1.2.52SIP Status: 100 Giving a try
  0.00183010.1.1.82 -> 10.1.1.60SIP/SDP Request: INVITE
sip:18148656...@10.1.1.60, with session description
  0.00434310.1.1.60 -> 10.1.1.82SIP Status: 100 Trying
  1.47410010.1.1.60 -> 10.1.1.82SIP/SDP Status: 183 Session
Progress, with session description
  1.47424010.1.1.82 -> 10.1.2.52SIP/SDP Status: 183 Session
Progress, with session description
  1.48385010.1.1.60 -> 10.1.1.82SIP Request: NOTIFY
sip:10.1.2.52:5060
  1.48396210.1.1.82 -> 10.1.2.52SIP Request: NOTIFY
sip:10.1.2.52:5060
  1.48609110.1.2.52 -> 10.1.1.82SIP Status: 200 OK
  1.48613310.1.1.82 -> 10.1.1.60SIP Status: 200 OK
  4.25237310.1.1.60 -> 10.1.1.82SIP/SDP Status: 200 OK, with session
description
  4.25275610.1.1.82 -> 10.1.2.52SIP/SDP Status: 200 OK, with session
description



On 1/12/11 6:19 AM, "Bogdan-Andrei Iancu"  wrote:

> Hi Sven,
> 
> "Bogus Event 8 in state 2" is translated to receiving an indialog
> request (non ACK, non BYE) while dialog in early state.maybe it is a
> PRACK to the 183can you check that ?
> 
> Regards,
> Bogdan
> 
> Sven Schulz wrote:
>> Running opensips 1.6.3, dialog module seems to function correctly
>> however I keep getting these messages:
>> 
>> CRITICAL:dialog:log_next_state_dlg: bogus event 8 in state 2 for dlg
>> 0x2b33956052c0 [2284:61359203] with clid
>> '11a0bd80-d2c160b0-1a-34020...@10.1.2.52
>> <%2711a0bd80-d2c160b0-1a-34020...@10.1.2.52>' and tags
>> 'd5edda48-9c10-424c-b200-8ec1eb8e532c-42504961' '292F5834-D13'
>> 
>> They only seem to happen when an INVITE is followed by a 183 RINGING
>> message. INVITES without a 183 wont get this error messege. Is this
>> normal or should I be concerned?
>> 
>> 
>> Sven Schulz
>> Penn State University
>> Telecommunications and Network Services
>> 814.865.6116
>> sip:s...@psu.edu
>> 
>> 
>> 
>> ___
>> Users mailing list
>> Users@lists.opensips.org
>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>   
> 


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] new install and INVITES not being forwarded

[Bogdan-Andrei Iancu]

Hi Gareth,

Gareth Blades wrote:
I am having a problem with running opensips in debug level 6. When 
opensips is set to this I am finding that it takes a long time to 
respond to register requests (over 5 seconds compared to a fraction of 
a second) which means that my phone times out when trying to register 
so I cannot then debug trying to send calls through.
woow...that is really strange.never encountered something like 
thatdo you log to stderror(console) or to syslog ?


Any ideas?

DNS timeouts are a common cause for these sort of pauses. Are there 
any specific DNS setup requirements that opensips has that I might 
have missing?

see the DNS related params:
   http://www.opensips.org/Resources/DocsCoreFcn#toc41

Also, better use DNS cache app between your opensips and the outer world.


So, your missing INVITE (not being sent out) are because of opensips 
blocking in some DNS query ?


Regards,
Bogdan


Thanks
Gareth

Bogdan-Andrei Iancu wrote:

Hi Gareth,

On a first look, the script looks ok, but as a general way to 
debug/troubleshoot your script, place xlog()'s in your script, in 
different points, to see if the script execution gets to that point.


Regarding the 408 - take care your script does not have NAT support, 
so it may not work if your client is behind a NAT.


Regards,
Bogdan

Gareth Blades wrote:

Thanks I will need to get back to you on Monday.

The phone on my desk has just started not being able to register. 
Its showing registration status 408 which is no response.
I can see the opensips server receiving the the registration and 
issuing a 401 unauthorised but I dont think this is getting back 
through the firewall.
I will get some static forwarding added to the firewall to avoid 
this sort of issue complicaing investigating the other problem.


In the meantime I have attached my config file.


Bogdan-Andrei Iancu wrote:

Hi Gareth,

looking at the logs, it seams that your script processing (for the 
INVITE) never get to a t_relay() point - place some xlog() prints 
in your script to see where the INVITE processing is going through.


Regards,
Bogdan

Gareth Blades wrote:
Thanks. I have attached a text file as its a bit long to paste in 
a message and there are long lines.


Looking through the log these couple of entries look significant.

DBG:auth:build_auth_hf: 'Proxy-Authenticate: Digest 
realm="vmopensips1.skycomuk.com", 
nonce="4d26db660001d69ce33126031746bdf4735e3f0f922d"

...
DBG:tm:matching_3261: RFC3261 transaction matching failed




Bogdan-Andrei Iancu wrote:

Hi Gareth,

hard to say without some logs - do you see errors in the opensips 
log ? if not, increase the debug level to 6 and post the logs 
corresponding the call.


Regards,
Bogdan

Gareth Blades wrote:

Setup:-
Server installed at a hosting facility with its owm public IP 
address.
A desk phone in the office with a couple of accounts configured 
registered to opensips through a NAT firewall.
The opensips server has been installed and configured using the 
guide in the Opensips PACT book together with database support 
(no multidomain).


The two lines on the phone are able to register fine and I can 
see them as being registered in the database and 'opensipsctrl 
ul show' displays what I would expect.


However when I call between the two lines and minitoring via 
ngrep I see the initial INVITE go out and then a '407 proxy 
authentication required' comes back and the phone sends the 
invite again with the authentication but opensips doesnt send 
any reply. The phone retransmits the packet a couple of times 
and displays call failed.


Any idea what may be wrong?


I know the call audio is unlikely to work but that will be the 
next step. I am just trying to get each step working as I go at 
the moment.



Thanks
Gareth

___






___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users




--
Bogdan-Andrei Iancu
OpenSIPS Event - expo, conf, social, bootcamp
2 - 4 February 2011, ITExpo, Miami,  USA
www.voice-system.ro


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Pacth rtpproxy

[Denis Putyato]

Hello, Razvan

 

“This is a problem since we would like to have a long period for call 
establishment” and what does it mean “call establishment” in such context?

 

From: users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] On Behalf Of Razvan Crainea
Sent: Wednesday, January 12, 2011 5:18 PM
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Pacth rtpproxy

 

Hello Denis,

RTPProxy is only used to detect the media timeout. If OpenSIPS receives a 
timeout notification on an unestablished call, it simply ignores it.
If you want to terminate the call when the callee doesn't answer you can use 
the tm module and set the "fr_inv_timer" parameter. You can get more details 
from:
http://www.opensips.org/html/docs/modules/devel/tm.html#id250344

Regards,
Razvan

On 01/12/2011 02:38 PM, Denis Putyato wrote: 

Razvan, I got rtpproxy from   
http://opensips.org/pub/rtpproxy/ as you wrote.

I started it using such cli command “/usr/local/rtpproxy1/bin/rtpproxy -u 
opensips -l 1.1.1.1 -s /var/run/rtpproxy.sock -T 80 -i -n /var/run/timer.sock  
-d INFO” and made test call.

Callee has been ringing during about 2 minutes and nothing happens at all. What 
I did wrong? 

 

P.S. I use such function in my script for rtp proxy “rtpproxy_offer("con");”

 

From: users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] On Behalf Of Razvan Crainea
Sent: Wednesday, January 12, 2011 3:14 PM
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Pacth rtpproxy

 

Hello Denis,

In the official release of RTPProxy, the timeout parameter (-T) controls both 
session establishment and rtp timeout. This is a problem since we would like to 
have a long period for call establishment, but a fast media timeout detection.
In the patched version of RTPProxy, the -W parameter allows you to specify a 
longer period for call establishment. If not set, it has the default value of 
60 seconds. 
If you decide not to use patched version of RTPProxy, the timeout notification 
will work, but you will have the same timeout in both situations.

Regards,
Razvan


On 01/12/2011 07:38 AM, Denis Putyato wrote: 

Hello Razvan,

 

“OpenSIPS shouldn't even try to terminate the call because it isn't established 
yet”

As I understand I just do not need to use –W key when starting rtpproxy, it 
does not work at all?

 

 

From: users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] On Behalf Of Razvan Crainea
Sent: Tuesday, January 11, 2011 6:49 PM
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Pacth rtpproxy

 

Hello Denis,

You are right, OpenSIPS shouldn't even try to terminate the call because it 
isn't established yet. I just added a small fix to solve this problem. Please 
update your code from svn to use this fix.
The RTPProxy patch was done against commit 
"600c80493793bafd2d69427bc22fcb43faad98c5". You can either get the RTPProxy 
from git, change it's branch and then apply the patch, or you can download an 
already patched version from http://opensips.org/pub/rtpproxy/.

Regards,
Razvan 

On 1/11/2011 2:19 PM, Denis Putyato wrote: 

Hello!

 

I try patch rtpproxy gotten from git. And there is such error during patching

 

patch < rtpproxy_timeout_notification.patch

patching file main.c

Hunk #1 succeeded at 70 (offset 2 lines).

Hunk #2 FAILED at 120.

Hunk #3 succeeded at 132 with fuzz 1 (offset 4 lines).

Hunk #4 succeeded at 211 with fuzz 2 (offset 4 lines).

Hunk #5 succeeded at 276 (offset 4 lines).

Hunk #6 succeeded at 742 with fuzz 2 (offset -26 lines).

Hunk #7 succeeded at 758 with fuzz 2 (offset -26 lines).

1 out of 7 hunks FAILED -- saving rejects to file main.c.rej

patching file rtpp_command.c

Hunk #1 FAILED at 795.

Hunk #2 FAILED at 888.

2 out of 2 hunks FAILED -- saving rejects to file rtpp_command.c.rej

patching file rtpp_defines.h

Hunk #1 FAILED at 95.

1 out of 1 hunk FAILED -- saving rejects to file rtpp_defines.h.rej

patching file rtpp_notify.c

 

rtpproxy_timeout_notification.patch is a patch for timeout notification which  
divide rtp timeout and session initiation timeout notification as said in

http://www.opensips.org/html/docs/modules/devel/nathelper.html#id249142

 

This patch I got from SVN version of latest Opensips. 

 
 
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

 

 
 
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users






-- 
Razvan Crainea
www.voice-system.ro
 
 
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users





-- 
Razvan Crainea
www.voice-system.ro
___
Users mailing list
Users@lists.opensips.org
http://lists.opensip

Re: [OpenSIPS-Users] nat_traversal samples?

[Jeff Chua]

On Wed, Jan 12, 2011 at 7:56 PM, Bogdan-Andrei Iancu
 wrote:
> Hi Jeff,
>
> The UDP versus TCP issue is for the communication between opensips and media
> relay ? If so, how comes you have a firewall between them ? you have
> opensips and the media relay in different networks ?

Bogdan,

I'm looking for something like this ...

SIP Client A -- OpenSIPS B -- VPN (only TCP/IP allowed) -- OpenSIPS C
-- SIP Client D

I read that with nat_travesal, I won't even need rtpproxy or mediaproxy.


Thanks,
Jeff

___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Pacth rtpproxy

[Razvan Crainea]

Hello Denis,

RTPProxy is only used to detect the media timeout. If OpenSIPS receives 
a timeout notification on an unestablished call, it simply ignores it.
If you want to terminate the call when the callee doesn't answer you can 
use the tm module and set the "fr_inv_timer" parameter. You can get more 
details from:

http://www.opensips.org/html/docs/modules/devel/tm.html#id250344

Regards,
Razvan

On 01/12/2011 02:38 PM, Denis Putyato wrote:


Razvan, I got rtpproxy from http://opensips.org/pub/rtpproxy/ as you 
wrote.


I started it using such cli command “/usr/local/rtpproxy1/bin/rtpproxy 
-u opensips -l 1.1.1.1 -s /var/run/rtpproxy.sock -T 80 -i -n 
/var/run/timer.sock  -d INFO” and made test call.


Callee has been ringing during about 2 minutes and nothing happens at 
all. What I did wrong?


P.S. I use such function in my script for rtp proxy 
“rtpproxy_offer("con");”


*From:*users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] *On Behalf Of *Razvan Crainea

*Sent:* Wednesday, January 12, 2011 3:14 PM
*To:* OpenSIPS users mailling list
*Subject:* Re: [OpenSIPS-Users] Pacth rtpproxy

Hello Denis,

In the official release of RTPProxy, the timeout parameter (-T) 
controls both session establishment and rtp timeout. This is a problem 
since we would like to have a long period for call establishment, but 
a fast media timeout detection.
In the patched version of RTPProxy, the -W parameter allows you to 
specify a longer period for call establishment. If not set, it has the 
default value of 60 seconds.
If you decide not to use patched version of RTPProxy, the timeout 
notification will work, but you will have the same timeout in both 
situations.


Regards,
Razvan


On 01/12/2011 07:38 AM, Denis Putyato wrote:

Hello Razvan,

“OpenSIPS shouldn't even try to terminate the call because it isn't 
established yet”


As I understand I just do not need to use –W key when starting 
rtpproxy, it does not work at all?


*From:*users-boun...@lists.opensips.org 
 
[mailto:users-boun...@lists.opensips.org] *On Behalf Of *Razvan Crainea

*Sent:* Tuesday, January 11, 2011 6:49 PM
*To:* OpenSIPS users mailling list
*Subject:* Re: [OpenSIPS-Users] Pacth rtpproxy

Hello Denis,

You are right, OpenSIPS shouldn't even try to terminate the call 
because it isn't established yet. I just added a small fix to solve 
this problem. Please update your code from svn to use this fix.
The RTPProxy patch was done against commit 
"600c80493793bafd2d69427bc22fcb43faad98c5". You can either get the 
RTPProxy from git, change it's branch and then apply the patch, or you 
can download an already patched version from 
http://opensips.org/pub/rtpproxy/.


Regards,
Razvan

On 1/11/2011 2:19 PM, Denis Putyato wrote:

Hello!

I try patch rtpproxy gotten from git. And there is such error during 
patching


patch < rtpproxy_timeout_notification.patch

patching file main.c

Hunk #1 succeeded at 70 (offset 2 lines).

Hunk #2 FAILED at 120.

Hunk #3 succeeded at 132 with fuzz 1 (offset 4 lines).

Hunk #4 succeeded at 211 with fuzz 2 (offset 4 lines).

Hunk #5 succeeded at 276 (offset 4 lines).

Hunk #6 succeeded at 742 with fuzz 2 (offset -26 lines).

Hunk #7 succeeded at 758 with fuzz 2 (offset -26 lines).

1 out of 7 hunks FAILED -- saving rejects to file main.c.rej

patching file rtpp_command.c

Hunk #1 FAILED at 795.

Hunk #2 FAILED at 888.

2 out of 2 hunks FAILED -- saving rejects to file rtpp_command.c.rej

patching file rtpp_defines.h

Hunk #1 FAILED at 95.

1 out of 1 hunk FAILED -- saving rejects to file rtpp_defines.h.rej

patching file rtpp_notify.c

rtpproxy_timeout_notification.patch is a patch for timeout 
notification which  divide rtp timeout and session initiation timeout 
notification as said in


http://www.opensips.org/html/docs/modules/devel/nathelper.html#id249142

This patch I got from SVN version of latest Opensips.

  
  
___

Users mailing list
Users@lists.opensips.org  
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

  
  
___

Users mailing list
Users@lists.opensips.org  
http://lists.opensips.org/cgi-bin/mailman/listinfo/users



--
Razvan Crainea
www.voice-system.ro  


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


--
Razvan Crainea
www.voice-system.ro

___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Pacth rtpproxy

[Denis Putyato]

Razvan, I got rtpproxy from   
http://opensips.org/pub/rtpproxy/ as you wrote.

I started it using such cli command “/usr/local/rtpproxy1/bin/rtpproxy -u 
opensips -l 1.1.1.1 -s /var/run/rtpproxy.sock -T 80 -i -n /var/run/timer.sock  
-d INFO” and made test call.

Callee has been ringing during about 2 minutes and nothing happens at all. What 
I did wrong? 

 

P.S. I use such function in my script for rtp proxy “rtpproxy_offer("con");”

 

From: users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] On Behalf Of Razvan Crainea
Sent: Wednesday, January 12, 2011 3:14 PM
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Pacth rtpproxy

 

Hello Denis,

In the official release of RTPProxy, the timeout parameter (-T) controls both 
session establishment and rtp timeout. This is a problem since we would like to 
have a long period for call establishment, but a fast media timeout detection.
In the patched version of RTPProxy, the -W parameter allows you to specify a 
longer period for call establishment. If not set, it has the default value of 
60 seconds. 
If you decide not to use patched version of RTPProxy, the timeout notification 
will work, but you will have the same timeout in both situations.

Regards,
Razvan


On 01/12/2011 07:38 AM, Denis Putyato wrote: 

Hello Razvan,

 

“OpenSIPS shouldn't even try to terminate the call because it isn't established 
yet”

As I understand I just do not need to use –W key when starting rtpproxy, it 
does not work at all?

 

 

From: users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] On Behalf Of Razvan Crainea
Sent: Tuesday, January 11, 2011 6:49 PM
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Pacth rtpproxy

 

Hello Denis,

You are right, OpenSIPS shouldn't even try to terminate the call because it 
isn't established yet. I just added a small fix to solve this problem. Please 
update your code from svn to use this fix.
The RTPProxy patch was done against commit 
"600c80493793bafd2d69427bc22fcb43faad98c5". You can either get the RTPProxy 
from git, change it's branch and then apply the patch, or you can download an 
already patched version from http://opensips.org/pub/rtpproxy/.

Regards,
Razvan 

On 1/11/2011 2:19 PM, Denis Putyato wrote: 

Hello!

 

I try patch rtpproxy gotten from git. And there is such error during patching

 

patch < rtpproxy_timeout_notification.patch

patching file main.c

Hunk #1 succeeded at 70 (offset 2 lines).

Hunk #2 FAILED at 120.

Hunk #3 succeeded at 132 with fuzz 1 (offset 4 lines).

Hunk #4 succeeded at 211 with fuzz 2 (offset 4 lines).

Hunk #5 succeeded at 276 (offset 4 lines).

Hunk #6 succeeded at 742 with fuzz 2 (offset -26 lines).

Hunk #7 succeeded at 758 with fuzz 2 (offset -26 lines).

1 out of 7 hunks FAILED -- saving rejects to file main.c.rej

patching file rtpp_command.c

Hunk #1 FAILED at 795.

Hunk #2 FAILED at 888.

2 out of 2 hunks FAILED -- saving rejects to file rtpp_command.c.rej

patching file rtpp_defines.h

Hunk #1 FAILED at 95.

1 out of 1 hunk FAILED -- saving rejects to file rtpp_defines.h.rej

patching file rtpp_notify.c

 

rtpproxy_timeout_notification.patch is a patch for timeout notification which  
divide rtp timeout and session initiation timeout notification as said in

http://www.opensips.org/html/docs/modules/devel/nathelper.html#id249142

 

This patch I got from SVN version of latest Opensips. 

 
 
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

 

 
 
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users





-- 
Razvan Crainea
www.voice-system.ro
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] new install and INVITES not being forwarded

[Gareth Blades]

I am having a problem with running opensips in debug level 6. When 
opensips is set to this I am finding that it takes a long time to 
respond to register requests (over 5 seconds compared to a fraction of a 
second) which means that my phone times out when trying to register so I 
cannot then debug trying to send calls through.


Any ideas?

DNS timeouts are a common cause for these sort of pauses. Are there any 
specific DNS setup requirements that opensips has that I might have missing?


Thanks
Gareth

Bogdan-Andrei Iancu wrote:

Hi Gareth,

On a first look, the script looks ok, but as a general way to 
debug/troubleshoot your script, place xlog()'s in your script, in 
different points, to see if the script execution gets to that point.


Regarding the 408 - take care your script does not have NAT support, so 
it may not work if your client is behind a NAT.


Regards,
Bogdan

Gareth Blades wrote:

Thanks I will need to get back to you on Monday.

The phone on my desk has just started not being able to register. Its 
showing registration status 408 which is no response.
I can see the opensips server receiving the the registration and 
issuing a 401 unauthorised but I dont think this is getting back 
through the firewall.
I will get some static forwarding added to the firewall to avoid this 
sort of issue complicaing investigating the other problem.


In the meantime I have attached my config file.


Bogdan-Andrei Iancu wrote:

Hi Gareth,

looking at the logs, it seams that your script processing (for the 
INVITE) never get to a t_relay() point - place some xlog() prints in 
your script to see where the INVITE processing is going through.


Regards,
Bogdan

Gareth Blades wrote:
Thanks. I have attached a text file as its a bit long to paste in a 
message and there are long lines.


Looking through the log these couple of entries look significant.

DBG:auth:build_auth_hf: 'Proxy-Authenticate: Digest 
realm="vmopensips1.skycomuk.com", 
nonce="4d26db660001d69ce33126031746bdf4735e3f0f922d"

...
DBG:tm:matching_3261: RFC3261 transaction matching failed




Bogdan-Andrei Iancu wrote:

Hi Gareth,

hard to say without some logs - do you see errors in the opensips 
log ? if not, increase the debug level to 6 and post the logs 
corresponding the call.


Regards,
Bogdan

Gareth Blades wrote:

Setup:-
Server installed at a hosting facility with its owm public IP 
address.
A desk phone in the office with a couple of accounts configured 
registered to opensips through a NAT firewall.
The opensips server has been installed and configured using the 
guide in the Opensips PACT book together with database support (no 
multidomain).


The two lines on the phone are able to register fine and I can see 
them as being registered in the database and 'opensipsctrl ul 
show' displays what I would expect.


However when I call between the two lines and minitoring via ngrep 
I see the initial INVITE go out and then a '407 proxy 
authentication required' comes back and the phone sends the invite 
again with the authentication but opensips doesnt send any reply. 
The phone retransmits the packet a couple of times and displays 
call failed.


Any idea what may be wrong?


I know the call audio is unlikely to work but that will be the 
next step. I am just trying to get each step working as I go at 
the moment.



Thanks
Gareth

___






___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Problem with load balancer module

[Bogdan-Andrei Iancu]

Hi Diego,

The bug seams to be in your callee device. Take a look at the 200 OK it 
sends:


U 192.168.2.165:5061 -> 192.168.2.165:5060
SIP/2.0 200 OK..From: "Your Long 
Name";tag=A46E9878A6B36612423768382DD6C758..To: 
;tag=843e7f0-a502a8c0-13c5-50022-1110-60a48c56-1110..Call-ID: 
d059e3f6f50cacfb33b4526b0a1ca...@192.168.2
.150..CSeq: 1 INVITE..Via: SIP/2.0/UDP 
192.168.2.165;branch=z9hG4bK7bcf.2d3c8236.0..Via: SIP/2.0/UDP 
192.168.2.150:5060;received=192.168.2.150;rport=5060;branch=z9hG4bKC115ED4230E704ED2956D13FC3999153..Record-Route: 
78A6B36612423768382DD6C758;did=08c.697623b5>..Contact: 
..Allow: INVITE, CANCEL, ACK, BYE, 
OPTIONS, INFO..Content-Type: application/sdp..Content-Length: 
210v=0..o=RBTAlerting 2844730 0 IN IP4 192.168.2.165..s=INTEL_
SIP_CCLIB..i=session information..c=IN IP4 192.168.2.165..t=0 0..m=audio 
49152 RTP/AVP 8 101..a=rtpmap:8 PCMA/8000..a=rtpmap:101 
telephone-event/8000..

#

The 200 OK is sent from 192.168.2.165:5061, but in contact it places 
sip:50257609...@192.168.2.165 -> the port is missing, so the sequential 
requests are going to the wrong destination (to 
sip:50257609...@192.168.2.165 which is actually the LB ,not the callee)


Regards,
Bogdan

Diego Barberio wrote:

Hi Bogdan,

Have you been able to take a look at the traces I sent?

Thanks
Diego

-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Diego Barberio
Sent: lunes, 10 de enero de 2011 12:51 p.m.
To: users@lists.opensips.org
Subject: [OpenSIPS-Users] Problem with load balancer module

Hi Bogdan,

Thank you for your prompt response. I'm sorry I couldn't send the trace
before but I had some problems with my network.
Also I've change the IP Address schema.
The call is originated from 192.168.2.150 to 192.168.2.165:5060 which is the
opensips address. Currently, the load balancer is configured to redirect the
calls to 192.168.2.165:5061 or 192.168.2.165:5062. In the call I'm sending
the INVITE was redirected to 5061.

The you will see that the ACK is not redirected to the destination.

Thanks
Diego

-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
Sent: viernes, 07 de enero de 2011 01:19 p.m.
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Problem with load balancer module

Hi Diego,

Could you post a SIP capture of a complete call (starting with INVITE) from
the opensips LB machine ?

Regards,
Bogdan

Diego Barberio wrote:
  

Diego Sebastián Barberio

www.redmondsoftware.com
+54 11 48153511 (Ext 143)


-Original Message-
From: Diego Barberio [mailto:diego.barbe...@redmondsoftware.com]
Sent: jueves, 06 de enero de 2011 03:50 p.m.
To: 'users@lists.opensips.org'
Subject: Problem with load balancer module

Hello,

I’m testing the Load Balancing module and I have a problem I can’t fix 
by myself. The INVITE message is routed correctly to one of the


destinations.
  
However the subsequent ACK and the BYE messages are not sent to the 
destinations.


I set up opensips to run only in the udp 5060 port. Then I have two 
identical applications: one running on port 5061 and the other on port 
5062, the 3 components are running in the same server which has a 
single IP

address: 192.168.1.195.
The application is very simple:
1. Receives the INVITE, starts streaming the RTP, and sends the OK
2. When the ACK is received injects some music in the streaming
3. Waits until de BYE is received. Then stops the streaming and

sends 
  

the OK.

This is the configuration of the load_balancer table:

mysql> select * from load_balancer;
++--++---++---
++--++---++---
++--++---++---
+
| id | group_id | dst_uri| resources | probe_mode |
description |
++--++---++---
++--++---++---
++--++---++---
+
|  1 |0 | sip:192.168.1.195:5061 | pstn=1|  0 |
|
|  2 |0 | sip:192.168.1.195:5062 | pstn=1|  0 |
|
++--++---++---
++--++---++---
++--++---++---
+
I've configured only one resource in each application because I'm just 
testing.


Finally, this is the configuration script, which is the one from the 
tutorial on the website:


debug=3
log_facility=LOG_LOCAL6

fork=yes
children=4

/* uncomment the following lines to enable debugging */
debug=6
fork=no
#log_stderror=yes

/* uncomment the next line to disable TCP (default on) */ 
disable_tcp=yes


port=5060

/* uncomm

Re: [OpenSIPS-Users] Pacth rtpproxy

[Razvan Crainea]

Hello Denis,

In the official release of RTPProxy, the timeout parameter (-T) controls 
both session establishment and rtp timeout. This is a problem since we 
would like to have a long period for call establishment, but a fast 
media timeout detection.
In the patched version of RTPProxy, the -W parameter allows you to 
specify a longer period for call establishment. If not set, it has the 
default value of 60 seconds.
If you decide not to use patched version of RTPProxy, the timeout 
notification will work, but you will have the same timeout in both 
situations.


Regards,
Razvan


On 01/12/2011 07:38 AM, Denis Putyato wrote:


Hello Razvan,

“OpenSIPS shouldn't even try to terminate the call because it isn't 
established yet”


As I understand I just do not need to use –W key when starting 
rtpproxy, it does not work at all?


*From:*users-boun...@lists.opensips.org 
[mailto:users-boun...@lists.opensips.org] *On Behalf Of *Razvan Crainea

*Sent:* Tuesday, January 11, 2011 6:49 PM
*To:* OpenSIPS users mailling list
*Subject:* Re: [OpenSIPS-Users] Pacth rtpproxy

Hello Denis,

You are right, OpenSIPS shouldn't even try to terminate the call 
because it isn't established yet. I just added a small fix to solve 
this problem. Please update your code from svn to use this fix.
The RTPProxy patch was done against commit 
"600c80493793bafd2d69427bc22fcb43faad98c5". You can either get the 
RTPProxy from git, change it's branch and then apply the patch, or you 
can download an already patched version from 
http://opensips.org/pub/rtpproxy/.


Regards,
Razvan

On 1/11/2011 2:19 PM, Denis Putyato wrote:

Hello!

I try patch rtpproxy gotten from git. And there is such error during 
patching


patch < rtpproxy_timeout_notification.patch

patching file main.c

Hunk #1 succeeded at 70 (offset 2 lines).

Hunk #2 FAILED at 120.

Hunk #3 succeeded at 132 with fuzz 1 (offset 4 lines).

Hunk #4 succeeded at 211 with fuzz 2 (offset 4 lines).

Hunk #5 succeeded at 276 (offset 4 lines).

Hunk #6 succeeded at 742 with fuzz 2 (offset -26 lines).

Hunk #7 succeeded at 758 with fuzz 2 (offset -26 lines).

1 out of 7 hunks FAILED -- saving rejects to file main.c.rej

patching file rtpp_command.c

Hunk #1 FAILED at 795.

Hunk #2 FAILED at 888.

2 out of 2 hunks FAILED -- saving rejects to file rtpp_command.c.rej

patching file rtpp_defines.h

Hunk #1 FAILED at 95.

1 out of 1 hunk FAILED -- saving rejects to file rtpp_defines.h.rej

patching file rtpp_notify.c

rtpproxy_timeout_notification.patch is a patch for timeout 
notification which  divide rtp timeout and session initiation timeout 
notification as said in


http://www.opensips.org/html/docs/modules/devel/nathelper.html#id249142

This patch I got from SVN version of latest Opensips.

  
  
___

Users mailing list
Users@lists.opensips.org  
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


--
Razvan Crainea
www.voice-system.ro

___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Problem with load balancer module

[Diego Barberio]

Hi Bogdan,

Have you been able to take a look at the traces I sent?

Thanks
Diego

-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Diego Barberio
Sent: lunes, 10 de enero de 2011 12:51 p.m.
To: users@lists.opensips.org
Subject: [OpenSIPS-Users] Problem with load balancer module

Hi Bogdan,

Thank you for your prompt response. I'm sorry I couldn't send the trace
before but I had some problems with my network.
Also I've change the IP Address schema.
The call is originated from 192.168.2.150 to 192.168.2.165:5060 which is the
opensips address. Currently, the load balancer is configured to redirect the
calls to 192.168.2.165:5061 or 192.168.2.165:5062. In the call I'm sending
the INVITE was redirected to 5061.

The you will see that the ACK is not redirected to the destination.

Thanks
Diego

-Original Message-
From: users-boun...@lists.opensips.org
[mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu
Sent: viernes, 07 de enero de 2011 01:19 p.m.
To: OpenSIPS users mailling list
Subject: Re: [OpenSIPS-Users] Problem with load balancer module

Hi Diego,

Could you post a SIP capture of a complete call (starting with INVITE) from
the opensips LB machine ?

Regards,
Bogdan

Diego Barberio wrote:
> Diego Sebastián Barberio
>
> www.redmondsoftware.com
> +54 11 48153511 (Ext 143)
>
>
> -Original Message-
> From: Diego Barberio [mailto:diego.barbe...@redmondsoftware.com]
> Sent: jueves, 06 de enero de 2011 03:50 p.m.
> To: 'users@lists.opensips.org'
> Subject: Problem with load balancer module
>
> Hello,
>
> I’m testing the Load Balancing module and I have a problem I can’t fix 
> by myself. The INVITE message is routed correctly to one of the
destinations.
> However the subsequent ACK and the BYE messages are not sent to the 
> destinations.
>
> I set up opensips to run only in the udp 5060 port. Then I have two 
> identical applications: one running on port 5061 and the other on port 
> 5062, the 3 components are running in the same server which has a 
> single IP
> address: 192.168.1.195.
> The application is very simple:
>   1. Receives the INVITE, starts streaming the RTP, and sends the OK
>   2. When the ACK is received injects some music in the streaming
>   3. Waits until de BYE is received. Then stops the streaming and
sends 
> the OK.
>
> This is the configuration of the load_balancer table:
>
> mysql> select * from load_balancer;
> ++--++---++---
> ++--++---++---
> ++--++---++---
> +
> | id | group_id | dst_uri| resources | probe_mode |
> description |
> ++--++---++---
> ++--++---++---
> ++--++---++---
> +
> |  1 |0 | sip:192.168.1.195:5061 | pstn=1|  0 |
> |
> |  2 |0 | sip:192.168.1.195:5062 | pstn=1|  0 |
> |
> ++--++---++---
> ++--++---++---
> ++--++---++---
> +
> I've configured only one resource in each application because I'm just 
> testing.
>
> Finally, this is the configuration script, which is the one from the 
> tutorial on the website:
>
> debug=3
> log_facility=LOG_LOCAL6
>
> fork=yes
> children=4
>
> /* uncomment the following lines to enable debugging */
> debug=6
> fork=no
> #log_stderror=yes
>
> /* uncomment the next line to disable TCP (default on) */ 
> disable_tcp=yes
>
> port=5060
>
> /* uncomment and configure the following line if you want opensips to 
>bind on a specific interface/port/proto (default bind on all
> available) */ listen=udp:192.168.1.195:5060
>
> ### Modules Section 
>
> #set module path
> mpath="/usr/local/lib/opensips/modules/"
>
> loadmodule "maxfwd.so"
> loadmodule "sl.so"
> loadmodule "db_mysql.so"
> loadmodule "tm.so"
> loadmodule "uri.so"
> loadmodule "rr.so"
> loadmodule "dialog.so"
> loadmodule "mi_fifo.so"
> loadmodule "signaling.so"
> loadmodule "textops.so"
> loadmodule "load_balancer.so"
>
> # - setting module-specific parameters --- 
> # - mi_fifo params - modparam("mi_fifo", "fifo_name",
> "/tmp/opensips_fifo")
>
>
> # - rr params -
> # add value to ;lr param to cope with most of the UAs #modparam("rr", 
> "enable_full_lr", 1) # do not append from tag to the RR (no need for 
> this script) #modparam("rr", "append_fromtag", 0)
> modparam("rr","enable_double_rr",1)
> modparam("rr","append_fromtag",1)
>
>
> # - uri params -
> modparam("uri", "use_uri_table", 0)
>
> modparam("dialog", "dlg_flag", 13)
> modpa

Re: [OpenSIPS-Users] Prepay without RADIUS ?

[Bogdan-Andrei Iancu]

Hi Bogdan

more or less you need a fresh start - for example the default script is 
using DB for auth and acc part. What you miss is the authorize for PP 
(checking some balance or something like that) that can be done via some 
stored procedure in Mysql - and from script you run that procedure to 
see if the caller is allowed to call the destination and for how 
long...And your billing/rating will be hidden behind that DB procedure


Regards,
Bogdan

rad bogdan wrote:

Hi Bogdan,

Thank you for the response.
I've searched on the internet for some guidelines on how to configure 
the system with a DB backend but I wasn't able to find something 
useful. Can you tell me briefly how to start the configuration, 
considering that callcontrol, cdrtool, opensips and mediaproxy are 
already configured. The configuration files of all applications have 
references to RADIUS server. Can I simply ignore them and leave the 
default ones when working with the DB backend ?


Thanks,
Bogdan

--- On *Fri, 1/7/11, Bogdan-Andrei Iancu //* 
wrote:



From: Bogdan-Andrei Iancu 
Subject: Re: [OpenSIPS-Users] Prepay without RADIUS ?
To: "OpenSIPS users mailling list" 
Date: Friday, January 7, 2011, 6:33 PM

Hi Bogdan,

you can use a DB backend (instead of RADIUS) or an external
application to communicate with opensips (like via XMLRPC).

Regards,
Bogdan

rad bogdan wrote:
> Hi everyone,
>  Is it possible to configure a prepay system that doesn't use
the RADIUS server but only OpenSIPS, CallControl and CDRTool ?
> Maybe I haven't searched enough but the documentations that I've
read on internet rely on Freeradius being configured for
authentication-accounting-authorization and don't give other
alternatives.
> Couldn't OpenSIPS be used to accomplish all these tasks ?
>  Thanks,
> Bogdan
>
>
>

>
> ___
> Users mailing list
> Users@lists.opensips.org 
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>   



-- Bogdan-Andrei Iancu
OpenSIPS Event - expo, conf, social, bootcamp
2 - 4 February 2011, ITExpo, Miami,  USA
www.voice-system.ro


___
Users mailing list
Users@lists.opensips.org 
http://lists.opensips.org/cgi-bin/mailman/listinfo/users




___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
  



--
Bogdan-Andrei Iancu
OpenSIPS Event - expo, conf, social, bootcamp
2 - 4 February 2011, ITExpo, Miami,  USA
www.voice-system.ro


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] nat_traversal samples?

[Bogdan-Andrei Iancu]

Hi Jeff,

The UDP versus TCP issue is for the communication between opensips and 
media relay ? If so, how comes you have a firewall between them ? you 
have opensips and the media relay in different networks ?


Regards,
Bogdan

Jeff Chua wrote:

On Mon, Jan 10, 2011 at 8:09 PM, Bogdan-Andrei Iancu
 wrote:
  

Hi Jeff,

You can see how nat traversal is done with nathelper + RTPproxy - download
the opensips virtual machine
(http://www.voice-system.ro/shortcuts::opensips_livedvd)  were you have a
ready to run opensips platform with NAT traversal support - you can see in
the script form the VM how the NAT traversal is done (for signalling and
media).

Regards,
Bogdan



Bogdan,

Thanks for the pointer. I've just downloaded the vm, and looked thru
the code. It's using nathelper with rtpproxy ... but rtpproxy only
support UDP. I'm using VPN from my home to my office and the firewall
doesn't allow UDP. That's why I'm thinking of using nat_travesal which
supports TCP. Any pointer to sample config with nat_travesal?

Thanks,
Jeff

___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

  



--
Bogdan-Andrei Iancu
OpenSIPS Event - expo, conf, social, bootcamp
2 - 4 February 2011, ITExpo, Miami,  USA
www.voice-system.ro


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] b2b refer scenario

[Anton Zagorskiy]

I'm trying to solve accounting + lines limit (numbers of incoming and
outgoing calls) + refer problem..
The problem seems as very hard to me.

Everything works well, but refer request. 

There are 3 (A, B, C) UAs, each has limititation of number of incoming and
outgoing calls.
1. A calls B. Now A takes 1 outgoing and 0 incoming lines. Lets write this
as (1;0). So, B takes (0;1)
2. B decides to make an attended refer to C. So, it calls C (sent INVITE).
Now B takes (1;1), C - (0;1)
3. After conversation with C, B finishs the refer and send REFER to A.

For a properly billing and lines limit I should have :
1. Lines limit: A(1;0), B(1;1), C(0;1)
2. Conversations between A,B and B,C (from billing points of view)

But after when A received REFER request, it 
1. sends an INVITE, not a RE-INVITE, because of REFER has 'reffered-to;
header and 'replaces' tag, so that INVITE doesn't have 'to-tag', so this is
a new session with new billing between A and C.
2. sends a BYE to B, so this finishes the session between A and B and
changes line limit: now B takes (0;1)

After when C received that INVITE, he sends a BYE to B.

And now B takes 0 incoming and 0 outgoing lines and accounting (billing)
doesn't work properly (there is just one leg between A and C)


I haven't any idea how to solve this situation.
B2B could help me by means of sending RE-INVITE to A and C when B sends
REFER, instead of transmitting REFER to A.



WBR, Anton Zagorskiy
VoIP Developer, Oyster Telecom
Phone.: +7 812 601-0666
Fax: +7 812 601-0593
a.zagors...@oyster-telecom.ru
www.oyster-telecom.ru



> -Original Message-
> From: users-boun...@lists.opensips.org [mailto:users-
> boun...@lists.opensips.org] On Behalf Of Anca Vamanu
> Sent: Tuesday, January 11, 2011 6:59 PM
> To: users@lists.opensips.org
> Subject: Re: [OpenSIPS-Users] b2b refer scenario
> 
> On 01/11/2011 04:56 PM, Anton Zagorskiy wrote:
> > Hi Anca.
> >
> > As I see from the b2b documentation, there is no easy way to catch an
> > attended refer and make it as INVITE.
> Why do you want to do that?
> 
> >   The problem is in the refer-to field
> > that contains "Replaces" tag.
> >
> >
> > May be anyone has an idea how to do this?
> >
> >
> >
> >
> > Also, why there isn't documnetation for 1.6.4 on the site? There are
> 1.6.3
> > and trunk/devel in Cookbooks and 1.6.x and 1.7.x in the tutorials.
> What is
> > 1.7.x?
> >
> >
> >
> >
> > WBR, Anton Zagorskiy
> > VoIP Developer, Oyster Telecom
> > Phone.: +7 812 601-0666
> > Fax: +7 812 601-0593
> > a.zagors...@oyster-telecom.ru
> > www.oyster-telecom.ru
> >
> >
> >
> >> -Original Message-
> >> From: users-boun...@lists.opensips.org [mailto:users-
> >> boun...@lists.opensips.org] On Behalf Of Anca Vamanu
> >> Sent: Wednesday, January 05, 2011 3:53 PM
> >> To: users@lists.opensips.org
> >> Subject: Re: [OpenSIPS-Users] b2b refer scenario
> >>
> >> Hi Anton,
> >>
> >> 01/05/2011 02:09 PM, a.zagors...@oyster-telecom.ru wrote:
> >>> Hi Anca,
> >>> thanks for your reply, but could you explain a bit more?
> >>>
> >>> I don't understand how B2B will understand that he should execute
> >>> scenario on REFER request when b2b_init_request is calling on an
> >>> INVITE request?
> >>>
> >>> I'm talking about situation, where there are 2 UAs, they has
> >>> established session, and one of them sending the REFER request to
> >>> another.
> >>>
> >> Yes, this is exactly how it works, by calling the b2b_init_request
> on
> >> the initial Invite, so that B2BUA is in the middle of the call and
> can
> >> control it. If you look in the xml scenario file you will see that
> >> there
> >> is a rule for REFER there - this is how it know that when a REFER is
> >> received it has to do what it has to do.
> >> Try this out  and you will see ;) .
> >>
> >> Regards,
> >>
> >> --
> >> Anca Vamanu
> >> www.voice-system.ro
> >>
> >>
> >> ___
> >> Users mailing list
> >> Users@lists.opensips.org
> >> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
> >
> > ___
> > Users mailing list
> > Users@lists.opensips.org
> > http://lists.opensips.org/cgi-bin/mailman/listinfo/users
> >
> 
> 
> --
> Anca Vamanu
> www.voice-system.ro
> 
> 
> ___
> Users mailing list
> Users@lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Dialog Module and Bogus Event 8 in state 2

[Bogdan-Andrei Iancu]

Hi Sven,

"Bogus Event 8 in state 2" is translated to receiving an indialog 
request (non ACK, non BYE) while dialog in early state.maybe it is a 
PRACK to the 183can you check that ?


Regards,
Bogdan

Sven Schulz wrote:
Running opensips 1.6.3, dialog module seems to function correctly 
however I keep getting these messages:


CRITICAL:dialog:log_next_state_dlg: bogus event 8 in state 2 for dlg 
0x2b33956052c0 [2284:61359203] with clid 
'11a0bd80-d2c160b0-1a-34020...@10.1.2.52 
<%2711a0bd80-d2c160b0-1a-34020...@10.1.2.52>' and tags 
'd5edda48-9c10-424c-b200-8ec1eb8e532c-42504961' '292F5834-D13'


They only seem to happen when an INVITE is followed by a 183 RINGING 
message. INVITES without a 183 wont get this error messege. Is this 
normal or should I be concerned?



Sven Schulz
Penn State University
Telecommunications and Network Services
814.865.6116
sip:s...@psu.edu



___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
  



--
Bogdan-Andrei Iancu
OpenSIPS Event - expo, conf, social, bootcamp
2 - 4 February 2011, ITExpo, Miami,  USA
www.voice-system.ro


___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] nat_traversal samples?

[Jeff Chua]

On Mon, Jan 10, 2011 at 8:09 PM, Bogdan-Andrei Iancu
 wrote:
> Hi Jeff,
>
> You can see how nat traversal is done with nathelper + RTPproxy - download
> the opensips virtual machine
> (http://www.voice-system.ro/shortcuts::opensips_livedvd)  were you have a
> ready to run opensips platform with NAT traversal support - you can see in
> the script form the VM how the NAT traversal is done (for signalling and
> media).
>
> Regards,
> Bogdan

Bogdan,

Thanks for the pointer. I've just downloaded the vm, and looked thru
the code. It's using nathelper with rtpproxy ... but rtpproxy only
support UDP. I'm using VPN from my home to my office and the firewall
doesn't allow UDP. That's why I'm thinking of using nat_travesal which
supports TCP. Any pointer to sample config with nat_travesal?

Thanks,
Jeff

___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users