[OpenSIPS-Users] ERROR:registrar:update_contacts: invalid cseq for aor

[Nabeel]

Hi,

Getting a lot of these errors lately which has caused OpenSIPS to crash:

*ERROR:registrar:update_contacts: invalid cseq for aor XXX*


 I have added t_newtran() before save(location) in the config but the error
still seems to occur.

Nabeel

<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free.
www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Error: invalid sendtoparameters

[Nabeel]

Hi,

The listen part parameter is as follows:

listen=udp:162.XXX.X.206:5060


OpenSIPS version: 2.2

OS: CentOS 7



On 20 October 2017 at 08:50, Abdul Basit  wrote:

> please share you opnesips.cfg especially the section you configured listen
> parameter.
>
> what's opensips version and OS you are using?
>
> --
> regards,
>
> abdul basit | p: +92 32 1416 4196 | o: +92 30 0841 1445
>
> On 20 October 2017 at 09:22, Nabeel  wrote:
>
>> Hello,
>>
>> Please help me understand this error below. Is it of concern?
>> The server is not bound to localhost, it is bound to public IPs.
>>
>>
>>> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
>>> ERROR:core:proto_udp_send: 
>>> sendto(sock,0x7f1f89c2bcf0,4,0,0x7ffc5f146db0,16):
>>> Invalid argument(22)
>>> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
>>> CRITICAL:core:proto_udp_send: invalid sendtoparameters#012one possible
>>> reason is the server is bound to localhost and#012attempts to send to the
>>> net
>>> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
>>> ERROR:nathelper:msg_send: send() for proto 1 failed
>>> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
>>> ERROR:nathelper:nh_timer: sip msg_send failed!
>>
>>
>>
>> ___
>> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] Error: invalid sendtoparameters

[Nabeel]

Hello,

Please help me understand this error below. Is it of concern?
The server is not bound to localhost, it is bound to public IPs.


> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
> ERROR:core:proto_udp_send:
> sendto(sock,0x7f1f89c2bcf0,4,0,0x7ffc5f146db0,16): Invalid argument(22)
> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
> CRITICAL:core:proto_udp_send: invalid sendtoparameters#012one possible
> reason is the server is bound to localhost and#012attempts to send to the
> net
> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
> ERROR:nathelper:msg_send: send() for proto 1 failed
> Oct 20 05:10:12 server2 /usr/local/sbin/opensips[923]:
> ERROR:nathelper:nh_timer: sip msg_send failed!
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS winning the Google Open Source Peer Bonus

[Nabeel]

Congrats to OpenSIPS.

On 9 October 2017 at 12:05, Bogdan-Andrei Iancu  wrote:

>
> We are all proud to announce that the OpenSIPS project is a winner of the
> Google Open Source Peer Bonus - this is an official recognition from Google
> in terms of the OSS they use.
>
> "We’re excited to announce 2017’s second round of Open Source Peer Bonus
> winners. Google Open Source  established
> this program six years ago to encourage Googlers to recognize and celebrate
> the external developers contributing to the open source ecosystem Google
> depends on."
>
> https://opensource.googleblog.com/2017/10/more-open-source-
> peer-bonus-winners.html
>
> Thank you Google,
>
> --
> Bogdan-Andrei Iancu
>   OpenSIPS Founder and Developer
>   http://www.opensips-solutions.com
>
>
> ___
> 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] Error in TURN Servers

[Nabeel]

The screenshot doesn't show all the processes but I know it is hundreds
because it's necessary to run 'killall' or similar command several times to
get rid of all the processes and I can see the PIDs change each time.

I think it absolutely can be OpenSIPS server causing the TURN error because
OpenSIPS is in charge of creating the sessions in the first place, and if
there are hundreds of OpenSIPS processes running then there will be
hundreds of sessions created for each call.


On 10 August 2017 at 13:12, Max Mühlbronner 
wrote:

> Hi,
>
>
> I don't see hundreds or even thousands on the screenshot?
>
>
> Also the Opensips server should not have any connection/relation to your
> TURN Server, so i don't think Opensips could be the issue.
>
> A quick google search for "TURN 437 Allocation Mismatch" suggests that
> it's a TURN CLIENT <--> TURN Server problem. It could be related to the
> TURN client implementation.
>
>
> BR
>
> Max M.
> --
> *Von:* Users  im Auftrag von Nabeel <
> nabeelshik...@gmail.com>
> *Gesendet:* Donnerstag, 10. August 2017 13:55
> *An:* OpenSIPS users mailling list
> *Betreff:* Re: [OpenSIPS-Users] Error in TURN Servers
>
> Anyone? I am using OpenSIPS version 2.2.
>
> On 10 Aug 2017 8:02 am, "Nabeel"  wrote:
>
>> I found the cause of this. My OpenSIPS server has gone ballistic and is
>> running hundreds of processes/instances concurrently; please see this
>> screenshot: https://www.dropbox.com/s/7o7fsuab1jggrsh/Screen
>> shot%202017-08-10%2007.28.15.png?dl=0
>>
>> Even after I kill all the processes, hundreds of processes start again
>> whenever I start OpenSIPS in any way.
>>
>> Below is my opensips.service file but the same occurs if starting
>> OpenSIPS using 'opensipsctl start':
>>
>>
>> [Unit]
>>> Description=OpenSIPS is a very fast and flexible SIP (RFC3261) server
>>> After=network.target mariadb.service multi-user.target
>>
>>
>>
>>
>>
>> [Service]
>>> Type=forking
>>> User=root
>>> Group=root
>>> EnvironmentFile=-/etc/default/opensips
>>> PIDFile=/var/run/opensips.pid
>>> ExecStart=/usr/local/sbin/opensips -P /var/run/opensips.pid -f
>>> /usr/local/etc/opensips/opensips.cfg $OPTIONS
>>> ExecStartPre=/usr/local/sbin/opensips -c -f
>>> /usr/local/etc/opensips/opensips.cfg
>>> Restart=always
>>> TimeoutStopSec=30s
>>> LimitNOFILE=1048576
>>> LimitNPROC=1048576
>>>
>>
>>
>> [Install]
>>> WantedBy=multi-user.target
>>
>>
>>
>> Please advise how to fix this urgently.
>>
>> Nabeel
>>
>
> ___
> 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] Error in TURN Servers

[Nabeel]

Anyone? I am using OpenSIPS version 2.2.

On 10 Aug 2017 8:02 am, "Nabeel"  wrote:

> I found the cause of this. My OpenSIPS server has gone ballistic and is
> running hundreds of processes/instances concurrently; please see this
> screenshot: https://www.dropbox.com/s/7o7fsuab1jggrsh/
> Screenshot%202017-08-10%2007.28.15.png?dl=0
>
> Even after I kill all the processes, hundreds of processes start again
> whenever I start OpenSIPS in any way.
>
> Below is my opensips.service file but the same occurs if starting OpenSIPS
> using 'opensipsctl start':
>
>
> [Unit]
>> Description=OpenSIPS is a very fast and flexible SIP (RFC3261) server
>> After=network.target mariadb.service multi-user.target
>
>
>
>
>
> [Service]
>> Type=forking
>> User=root
>> Group=root
>> EnvironmentFile=-/etc/default/opensips
>> PIDFile=/var/run/opensips.pid
>> ExecStart=/usr/local/sbin/opensips -P /var/run/opensips.pid -f
>> /usr/local/etc/opensips/opensips.cfg $OPTIONS
>> ExecStartPre=/usr/local/sbin/opensips -c -f /usr/local/etc/opensips/
>> opensips.cfg
>> Restart=always
>> TimeoutStopSec=30s
>> LimitNOFILE=1048576
>> LimitNPROC=1048576
>>
>
>
> [Install]
>> WantedBy=multi-user.target
>
>
>
> Please advise how to fix this urgently.
>
> Nabeel
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Error in TURN Servers

[Nabeel]

I found the cause of this. My OpenSIPS server has gone ballistic and is
running hundreds of processes/instances concurrently; please see this
screenshot:
https://www.dropbox.com/s/7o7fsuab1jggrsh/Screenshot%202017-08-10%2007.28.15.png?dl=0

Even after I kill all the processes, hundreds of processes start again
whenever I start OpenSIPS in any way.

Below is my opensips.service file but the same occurs if starting OpenSIPS
using 'opensipsctl start':


[Unit]
> Description=OpenSIPS is a very fast and flexible SIP (RFC3261) server
> After=network.target mariadb.service multi-user.target





[Service]
> Type=forking
> User=root
> Group=root
> EnvironmentFile=-/etc/default/opensips
> PIDFile=/var/run/opensips.pid
> ExecStart=/usr/local/sbin/opensips -P /var/run/opensips.pid -f
> /usr/local/etc/opensips/opensips.cfg $OPTIONS
> ExecStartPre=/usr/local/sbin/opensips -c -f
> /usr/local/etc/opensips/opensips.cfg
> Restart=always
> TimeoutStopSec=30s
> LimitNOFILE=1048576
> LimitNPROC=1048576
>


[Install]
> WantedBy=multi-user.target



Please advise how to fix this urgently.

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

[OpenSIPS-Users] Error in TURN Servers

[Nabeel]

Hi,

I am recently getting the following error, '437 Allocation Mismatch', in
multiple Turn servers I have tried. The same error occurs in Coturn and
reTurnServer on multiple physical server instances and when using all
versions of the client I am using, which never had this issue. The common
factor in these cases, as a possible cause, is the SIP server, so I am
hoping someone will be able to advise. In the log of OpenSIPS itself, there
is nothing specific regarding this error. The error appears in all Turn
server logs as below:



Coturn log: session 001001: realm <> user <>: incoming packet
> message processed, *error 437: Mismatched allocation: wrong transaction*



reTurnServer log: DEBUG | 20170809-074953.802 | reTurnServer | RETURN |
> 139914911389440 | RequestHandler.cxx:455 | Allocation request received:
> localTuple=[UDP 162.259.6.236:3478], remoteTuple=[UDP 188.29.165.172:29101]
> WARNING | 20170809-074953.802 | reTurnServer | RETURN | 139914911389440 |
> RequestHandler.cxx:462 | *Allocation requested but already exists.
> Sending 437.* Sender=[UDP 188.29.165.172:29101] STACK |
> 20170809-074953.802 | reTurnServer | RETURN | 139914911389440 |
> StunMessage.cxx:1462 | Encoding stun message: *STUN Error Response:
> Allocate, id* 42a4122120c25f25602a28d9acdceb44 STACK |
> 20170809-074953.802 | reTurnServer | RETURN | 139914911389440 |
> StunMessage.cxx:1505 | *Encoding ErrorCode: 4 number=37 reason=Allocation
> Mismatch*
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Ghost calls 1001

[Nabeel]

In case the call is attempted via your server, you can add the following to
opensips.cfg to block sip scanners:

 if($ua=~"friendly-scanner") {
xlog("L_ERROR", "Auth error for $fU@$fd from $si method $rm
user-agent (friendly-scanner)\n");
drop();
exit;
 }
  if($ua=~"sipvicious") {
xlog("L_ERROR", "Auth error for $fU@$fd from $si method $rm
user-agent (friendly-scanner)\n");
drop();
exit;
 }

On 21 Apr 2017 8:12 a.m., "Uzair Hassan"  wrote:

> Is there any documentation I could read to understand the process you just
> described?
>
> On April 20, 2017 11:15:54 PM Schneur Rosenberg 
> wrote:
>
>> In addition to iptables/fail2ban you should inspect the useragent that
>> the packets come from, most of them will come from sip vicious or friendly
>> scanner etc, you can block them with iptables and/or with drop() in
>> opensips, this will stop the scanner right away because he won't get any
>> replies so he will just move on.
>>
>> On Apr 21, 2017 8:11 AM, "Uzair Hassan"  wrote:
>>
>>> Is there a way to change opensips port ? Whenever I try it doesn't even
>>> start.
>>>
>>> On April 20, 2017 9:09:55 PM "Alexander Jankowsky" <
>>> e75a4...@exemail.com.au> wrote:
>>>


 You might need to do a Wireshark trace and find out if the calls
 originate externally into the system.

 If you are in an open DMZ with the router, that could be just the start
 of your problems.

 I had Opensips 2.3.0-beta in the open on DMZ with the router for only a
 few hours and

 I then had a couple of dozen automated break in attempts trying to
 access the system.

 You need to pay a lot of attention to the system logs otherwise you may
 not even notice.

 Go over your router very carefully and restrict everything you do not
 need exposed.

 Port 5060 is a very popular target with automated robots, use another
 port if your able to.



 Alex





 *From:* Users [mailto:users-boun...@lists.opensips.org] *On Behalf Of 
 *Uzair
 Hassan
 *Sent:* Friday, 21 April 2017 6:16 AM
 *To:* users@lists.opensips.org
 *Subject:* [OpenSIPS-Users] Ghost calls 1001



 Hello all,



 I have setup a opensips 2.3 on a new server and I'm getting ghost calls
 into my system. How do I stop these ghost call? The opensips server is
 brand new. the install is clean and nothing has been touched after the
 initial simple residential script setup. What can I do to defend myself
 from these ghost calls.

 Thank you so much.


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

Re: [OpenSIPS-Users] Ghost calls 1001

[Nabeel]

In a ghost call, there is no RTP -- only the INVITE. If you answer a ghost
call, there will be no response. Usually, an IP scanner named 'SipVicious'
directly sends the INVITE to your client, so your server may not come into
play at all.

On 20 Apr 2017 10:32 p.m., "Mundkowsky, Robert"  wrote:

> Do you mean a client is using SIP/RTP to make a call direct to backend
> servers and bypassing the opensips proxy? Or somehow just using RTP without
> SIP to bypass the opensips proxy?
>
>
>
>
>
> Robert
>
>
>
> *From:* Users [mailto:users-boun...@lists.opensips.org] *On Behalf Of *
> Nabeel
> *Sent:* Thursday, April 20, 2017 5:17 PM
> *To:* OpenSIPS users mailling list 
> *Subject:* Re: [OpenSIPS-Users] Ghost calls 1001
>
>
>
> My understanding of ghost calls is that they go directly via the client
> through a loophole in the IP range rather than through the SIP server
> itself. In this case, server-based solutions don't seem likely to work?
>
>
>
> On 20 Apr 2017 10:08 p.m., "Mundkowsky, Robert" 
> wrote:
>
> User authentication at SIP level as well.
>
>
>
> Robert
>
>
>
> *From:* Users [mailto:users-boun...@lists.opensips.org] *On Behalf Of *Aqs
> Younas
> *Sent:* Thursday, April 20, 2017 4:55 PM
> *To:* OpenSIPS users mailling list 
> *Subject:* Re: [OpenSIPS-Users] Ghost calls 1001
>
>
>
> iptables, fail2ban and ip authentication if your users have static ips.
>
>
>
> On 21 April 2017 at 01:46, Uzair Hassan  wrote:
>
> Hello all,
>
>
>
> I have setup a opensips 2.3 on a new server and I'm getting ghost calls
> into my system. How do I stop these ghost call? The opensips server is
> brand new. the install is clean and nothing has been touched after the
> initial simple residential script setup. What can I do to defend myself
> from these ghost calls.
>
> Thank you so much.
>
>
>
>
> ___
> Users mailing list
> Users@lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
>
> --
>
> This e-mail and any files transmitted with it may contain privileged or
> confidential information. It is solely for use by the individual for whom
> it is intended, even if addressed incorrectly. If you received this e-mail
> in error, please notify the sender; do not disclose, copy, distribute, or
> take any action in reliance on the contents of this information; and delete
> it from your system. Any other use of this e-mail is prohibited.
>
>
>
> Thank you for your compliance.
> --
>
>
> ___
> Users mailing list
> Users@lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
> --
>
> This e-mail and any files transmitted with it may contain privileged or
> confidential information. It is solely for use by the individual for whom
> it is intended, even if addressed incorrectly. If you received this e-mail
> in error, please notify the sender; do not disclose, copy, distribute, or
> take any action in reliance on the contents of this information; and delete
> it from your system. Any other use of this e-mail is prohibited.
>
> Thank you for your compliance.
> --
>
> ___
> 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] Ghost calls 1001

[Nabeel]

My understanding of ghost calls is that they go directly via the client
through a loophole in the IP range rather than through the SIP server
itself. In this case, server-based solutions don't seem likely to work?

On 20 Apr 2017 10:08 p.m., "Mundkowsky, Robert"  wrote:

> User authentication at SIP level as well.
>
>
>
> Robert
>
>
>
> *From:* Users [mailto:users-boun...@lists.opensips.org] *On Behalf Of *Aqs
> Younas
> *Sent:* Thursday, April 20, 2017 4:55 PM
> *To:* OpenSIPS users mailling list 
> *Subject:* Re: [OpenSIPS-Users] Ghost calls 1001
>
>
>
> iptables, fail2ban and ip authentication if your users have static ips.
>
>
>
> On 21 April 2017 at 01:46, Uzair Hassan  wrote:
>
> Hello all,
>
>
>
> I have setup a opensips 2.3 on a new server and I'm getting ghost calls
> into my system. How do I stop these ghost call? The opensips server is
> brand new. the install is clean and nothing has been touched after the
> initial simple residential script setup. What can I do to defend myself
> from these ghost calls.
>
> Thank you so much.
>
>
>
>
> ___
> Users mailing list
> Users@lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
> --
>
> This e-mail and any files transmitted with it may contain privileged or
> confidential information. It is solely for use by the individual for whom
> it is intended, even if addressed incorrectly. If you received this e-mail
> in error, please notify the sender; do not disclose, copy, distribute, or
> take any action in reliance on the contents of this information; and delete
> it from your system. Any other use of this e-mail is prohibited.
>
> Thank you for your compliance.
> --
>
> ___
> 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] Ghost calls 1001

[Nabeel]

Hi,

You can set client to use random port instead of standard 5060.

But a better way is to set the client to only allow your required domain,
if possible.

On 20 Apr 2017 9:51 p.m., "Uzair Hassan"  wrote:

Hello all,

I have setup a opensips 2.3 on a new server and I'm getting ghost calls
into my system. How do I stop these ghost call? The opensips server is
brand new. the install is clean and nothing has been touched after the
initial simple residential script setup. What can I do to defend myself
from these ghost calls.

Thank you so much.


___
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] Introducing OpenSIPS 2.3

[Nabeel]

Hi,

I run an ejabberd server for xmpp as well as OpenSIPS for SIP and was
wondering if these two could be integrated in any way in the future. These
servers run concurrently but they are completely independent of each other,
although for the same user base.

Nabeel

On 12 Jan 2017 4:48 p.m., "Bogdan-Andrei Iancu"  wrote:

> A new year has arrived, so it is the time for a new OpenSIPS major
> release – for OpenSIPS version 2.3 .
>
> For this version, the main focus on development is the *“integration”*,
> the integration of OpenSIPS with various external entities. Why is
> integration so important to end up being the main tag of a major release?
> Well, everybody in the VoIP world is operating VoIP platforms/systems – and
> these are more than SIP Engines (as OpenSIPS is). Indeed, the SIP Engine
> is the core and most important part of the platform, but to build something
> usable and useful, you need additional components into your platform like
> CDR/billing engines, monitoring and tracing tools, data backends, non-SIP
> trunking or more specialized SIP engines. Shortly you need your SIP Engine (
> OpenSIPS, of course) to be able to easily integrate with all these
> components.
>
> OpenSIPS 2.3  brings some new and exciting integration capabilities, that
> will definitely boost the value of your SIP platform:
>
>- extended Homer/SIPCapture <http://sipcapture.org/> integration to
>allow capturing of non-SIP data (transport level data, Management Interface
>commands, REST queries and more);
>- SIP-I support both in terms of passing-through and in terms of
>converting SIP-I to SIP and vice-versa ;
>- CGRates <http://cgrates.org/> integration for powerful
>rating/billing – everything in a simple and automatic way (via a dedicated
>module);
>- FreeSWITCH <https://freeswitch.org/> flavored Load-Balancing for a
>more realistic and accurate traffic balancing over FreeSWITCH clusters (as
>the load information is fetched in realtime from FreeSWITCH);
>- the Event Engine
><https://en.wikipedia.org/wiki/Event-driven_programming> to provide
>support for scenarios based on Subscribe/Notify model, where the script
>execution may subscribe and resume later according to certain events (like
>a dynamic implementation of the Push Notification mechanism);
>- extended RabbitMQ <https://www.rabbitmq.com/> support for custom and
>flexible data injection directly from OpenSIPS script;
>- extended Asynchronous support for more complex async scenarios (like
>launch with no wait);
>- more end-device integration (special SIP extensions).
>
> The timeline for OpenSIPS 2.3 is:
>
>- Beta Release – 13-17 March 2017
>- Stable Release – 24-28 April 2017
>- General Availability – 2nd of May 2017, during OpenSIPS Summit
><https://blog.opensips.org/2016/12/08/opensips-summit-2017-amsterdam/>
>
> To talk more about the features of this new release, a public audio
> conference <https://www.uberconference.com/opensips> will be available on 19th
> of January 2017, 4 pm GMT
> <https://www.timeanddate.com/worldclock/fixedtime.html?msg=Introducing+OpenSIPS+2.3&iso=20170119T18&p1=49&ah=1>,
>  thanks to
> the kind sponsorship of UberConference <https://www.uberconference.com/>.
> Anyone is welcome to join to find out more details or to ask questions about
> OpenSIPS 2.3 .
>
> This is a public and open conference, so no registration is needed, but if
> you want to announce your intention to participate, please let us know via the
> form on the blog post
> <https://blog.opensips.org/2017/01/12/introducing-opensips-2-3/>.
>
> Best regards,
>
> --
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
>
> ___
> 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] Merry Christmas and A Happy New Year

[Nabeel]

Merry Christmas to all.

Happy new year.

On 23 Dec 2016 4:11 p.m., "Bogdan-Andrei Iancu"  wrote:

> Hi all,
>
> As we all move one by one into the holiday mode, we - the entire OpenSIPS
> team - want to wish you Merry Christmas and A Happy New 2017. Thanks to all
> of you 2016 was a good year with many accomplishments and progress -
> nevertheless, let's make 2017 even greater. And we will start this by
> couple of public discussions about the OpenSIPS 2.3 release plan - a lot of
> interesting stuff there !
>
> Happy holidays,
>
> --
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developer
> http://www.opensips-solutions.com
>
>
> ___
> 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] Busy Signal When Using Voicemail

[Nabeel]

On 10 August 2016 at 19:38, Bogdan-Andrei Iancu  wrote:

> Hi Nabeel,
>
> OpenSIPS does not assume anything by default. If you want to have any new
> calls to user A rejected (if A already in a call, with other users or any
> service), you need to script this.
>


In the case of user A being in a call with other users, I didn't have to
script this in order to receive a 486 busy response.  I don't see this in
the cfg script. This seems to be default OpenSIPS behaviour in the case of
calling other users.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Busy Signal When Using Voicemail

[Nabeel]

Is there any way to make OpenSIPS handle a call with voicemail in exactly
the same way as a call with another SIP user?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Busy Signal When Using Voicemail

[Nabeel]

Hi,

The busy reply should be generated in exactly the same way as when the
callee is busy on the phone with another SIP user (from the OpenSIPS
subscriber table). The only difference is that the callee is on the phone
with asterisk voicemail instead of a SIP user from the subscriber table.
I'm not sure who exactly should generate the reply but I think it depends
on how OpenSIPS is interacting with the Voicemail server during a call to
voicemail. Perhaps it's not treating the voicemail service as another
'user' so the caller didn't receive the busy signal.

Nabeel

On 7 Aug 2016 6:57 p.m., "Bogdan-Andrei Iancu"  wrote:

> Hi Nabeel,
>
> Who should generate the 486 Busy reply ?
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 06.08.2016 02:56, Nabeel wrote:
>
> Hi,
>
> OpenSIPS does not receive a '486 Busy' signal when a callee is using the
> Asterisk voicemail service. If a user is currently listening to voicemail
> or leaving a voice message, an attempt to call that user should result in a
> '486 Busy' signal. From that response, I will be able to play a busy tone
> on the caller's side. Please let me know how to fix this.
>
> Nabeel
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] Busy Signal When Using Voicemail

[Nabeel]

Hi,

OpenSIPS does not receive a '486 Busy' signal when a callee is using the
Asterisk voicemail service. If a user is currently listening to voicemail
or leaving a voice message, an attempt to call that user should result in a
'486 Busy' signal. From that response, I will be able to play a busy tone
on the caller's side. Please let me know how to fix this.

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

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

Here is a list of changes I found:


1) Must build asterisk with ODBC storage enabled for voicemail because
using file storage will not store messages in the database.

2) Uncomment the lines *'odbcstorage=asterisk'* and
*'odbctable=voicemessages'* in voicemail.conf to enable database storage
for messages.

3) In file* /etc/asterisk/res_odbc.conf*, add the following under
[asterisk] to correctly limit database connection to the database:

limit => 5
share_connections => no
4) Change *'nat=yes'* to* 'nat=force_rport,comedia'* in the sipusers mysql
view because it is deprecated.

5) Add column 'callbackextension' to the sipusers mysql view using "*NULL
AS `callbackextension*" because it is required in the latest asterisk
version.

6) Change the priority ordering in extensions.conf to start with 1 instead
of n because the old syntax is deprecated.

7) Remove the suffix " |u " from extensions.conf because this causes
incorrect routing to mailboxes.

8) Add the line *'voicemail => odbc,asterisk,vmaliases'* to
extconfig.cfg because
this is missing in the current tutorial.

9) Remove the following parts from default opensips.cfg file because they
clash with the voicemail reply:

 if (!db_does_uri_exist()) {
>send_reply("420","Bad Extension");
>exit;
>    }
>


>t_newtran();
>t_reply("480", "Temporarily Unavailable");



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

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

I also found the correct way to deal with the LIMIT problem. Asterisk has a
built-in way to deal with this. In file* /etc/asterisk/res_odbc.conf*,  the
following should be added under [asterisk] :

limit => 5
share_connections => no

Now everything is working well without problems.

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

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

Hi Bogdan,

I have been able to solve that problem.
The issue was that I had asterisk compiled with file storage enabled
instead of ODBC storage. I recompiled asterisk with ODBC storage enabled
and now database storage is working.

Thanks.

Nabeel
On 14 Jul 2016 11:15 a.m., "Bogdan-Andrei Iancu" 
wrote:

> Hi Nabeel,
>
> That means the vmusers and vmaliases do work ok, still the VM storage
> engine does not. Do you have in voicemail.conf the following:
> odbcstorage=asteriskrt
> odbctable=voicemessages
>
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 05.07.2016 08:54, Nabeel wrote:
>
> I have been able to solve the issue of loading numbers without using the
> voicemail.conf file.
>
> After adding the line *'voicemail => odbc,asterisk,vmaliases'* to
> extconfig.cfg, I removed the suffix " |u " from extensions.conf:
>
> exten => _VMR_.,n,Voicemail(${EXTEN:4}*|u*)
>
>
>
> Now all phone numbers in the subscriber table are correctly linked to
> their mailboxes and I can successfully retrieve voice messages.
>
> I have set 'odbcstorage=asterisk' and 'odbctable=voicemessages' in
> voicemail.conf, but the 'voicemessages' table remains empty after leaving a
> message.
>
> Nabeel
>
>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

I have been able to solve the issue of loading numbers without using the
voicemail.conf file.

After adding the line *'voicemail => odbc,asterisk,vmaliases'* to
extconfig.cfg, I removed the suffix " |u " from extensions.conf:

exten => _VMR_.,n,Voicemail(${EXTEN:4}*|u*)



Now all phone numbers in the subscriber table are correctly linked to their
mailboxes and I can successfully retrieve voice messages.

I have set 'odbcstorage=asterisk' and 'odbctable=voicemessages' in
voicemail.conf, but the 'voicemessages' table remains empty after leaving a
message.

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

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

0.  The line block was in the default OpenSIPS config, but I agree that it
is not in the tutorial so should be removed (for voicemail).

1. I think there is a misunderstanding here.  'limit' is not a column; I am
referring to the mysql LIMIT clause:

https://dev.mysql.com/doc/refman/5.5/en/select.html

2. In the tutorial, the following is added to extconfig.cfg:

sipusers => odbc,asterisk,sipusers sippeers => odbc,asterisk,sipusers
voicemail => odbc,asterisk,vmusers meetme => odbc,asterisk,meetme

However, this is different to the config in your last Email with
'asteriskcfg'. Please clarify.

For the phone number +447479189410 in the subscriber table, I had to add
the following line to voicemail.conf for it to work (notice the " |u " ):

+447479189410|u => 1234,Example Mailbox,root@localhost

Asterisk seems to look for the " |u " suffix in addition to the  phone
number, instead of just the phone number which is in the subscriber table.

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

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

Hi Bogdan,

I just added the column to the view by adding "NULL AS `callbackextension`"
to the SQL view definition. I haven't linked the column to the subscriber
column, so this may not be the correct definition. However, it got rid of
the error.

About the voicemail.conf file, when I attempted to leave a voicemail to a
number in subscriber table, I got an error telling me that the user/mailbox
does not exist in voicemail.conf file. After adding that number manually to
the file, voicemail was working for that number. I think this may also mean
that the integration is not working correctly in my current setup.

Nabeel
On 4 Jul 2016 8:48 a.m., "Bogdan-Andrei Iancu"  wrote:

> Hi,
>
> What is the definition you used for this new column ?
>
> Best regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 02.07.2016 05:29, Nabeel wrote:
>
> In the last error message,* '**callbackextension = ?' *suggested that
> this column is missing from the sipusers mysql view. So I added this column
> to the view and now that error has been resolved. Only the following error
> remains now:
>
> [Jul  2 03:25:48] WARNING[19330][C-0005]: app.c:1633
>> __ast_play_and_record: No audio available on SIP/domain.com-0005??
>
>
>
>
>
>
> On 2 July 2016 at 02:36, Nabeel  wrote:
>
>> The issue in my last Email has solved the error about missing extension.
>> Now the following errors remain:
>>
>> [Jul  2 02:29:18] WARNING[18226][C-0001]: res_config_odbc.c:117
>>> custom_prepare: SQL Prepare failed![SELECT * FROM sipusers WHERE host = ?
>>> AND callbackextension = ? AND port = ?]
>>> [Jul  2 02:29:22] WARNING[18269][C-]: app.c:1633
>>> __ast_play_and_record: No audio available on SIP/domain.com-??
>>
>>
>>
>> On 2 July 2016 at 02:23, Nabeel  wrote:
>>
>>> The tutorial contains a mistake where the priority ordering in
>>> extensions.conf should start with 1, not n:
>>>
>>> ; Voicemail
>>>> exten => _VMR_.,1,Ringing
>>>> exten => _VMR_.,n,Wait(1)
>>>> exten => _VMR_.,n,Answer
>>>> exten => _VMR_.,n,Wait(1)
>>>> exten => _VMR_.,n,Voicemail(${EXTEN:4}|u)
>>>> exten => _VMR_.,n,Hangup
>>>
>>> ; Allow users to call their Voicemail directly
>>>> exten => VM_pickup,1,Ringing
>>>> exten => VM_pickup,n,wait(1)
>>>> exten => VM_pickup,n,VoicemailMain(${CALLERIDNUM}|s)
>>>> exten => VM_pickup,n,Hangup
>>>
>>>
>>>
>>>
>>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

Hi Samy,

Point 1 I cant imagine how those lines possibly relate to no media error in
> asterisk, I guess it depends on your config setup.


In point 1 I was referring to this error:

WARNING[17112] res_odbc.c: SetConnectAttr (Txn isolation) returned an
> error: HY000: [MySQL][ODBC 5.2(w) Driver]You have an error in your SQL
> syntax; check the manual that corresponds to your MariaDB server version
> for the right syntax to use near '7' at line 1


The above error was solved by adding "limit=1" to the sipusers mysql view.
Regarding the lines you are referring to -- removing them from the config
solved the 'no media error' because those reply messages were interfering
with the voicemail reply.

About your suggestion to use 'Asterisk Realtime', I thought that is
precisely what the tutorial is describing - how to integrate Asterisk
Realtime:

"This tutorial presents the concept and implementation of a realtime
integration of OpenSIPS SIP server and Asterisk media server."

I am a bit puzzled by your suggestion, but I will try asking in the
Asterisk mailing list.

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

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

The last error message has been solved by removing the following lines from
opensips.cfg:

 if (!db_does_uri_exist()) {
>send_reply("420","Bad Extension");
>exit;
>}
>


>t_newtran();
>t_reply("480", "Temporarily Unavailable");



Now voicemail seems to be working, but only if manually adding users to the
voicemail.conf file. So the following questions remain:

1. Adding "limit = 1" to sipusers mysql view resolves a database error. Is
this the correct way to fix the error for voicemail integration?

2. How can voicemail.conf file be configured to use variable substitution
for all users in the OpenSIPS subscriber table?

Nabeel

On 2 July 2016 at 13:41, Nabeel  wrote:

> In the latest version of Asterisk, there is a new file voicemail.conf
> which must be configured correctly for voicemail, but the tutorial does not
> mention this file at all. Please let me know how to configure this file for
> integration with OpenSIPS.
>
> Nabeel
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

In the latest version of Asterisk, there is a new file voicemail.conf which
must be configured correctly for voicemail, but the tutorial does not
mention this file at all. Please let me know how to configure this file for
integration with OpenSIPS.

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

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

In the last error message,* '**callbackextension = ?' *suggested that this
column is missing from the sipusers mysql view. So I added this column to
the view and now that error has been resolved. Only the following error
remains now:

[Jul  2 03:25:48] WARNING[19330][C-0005]: app.c:1633
> __ast_play_and_record: No audio available on SIP/domain.com-0005??






On 2 July 2016 at 02:36, Nabeel  wrote:

> The issue in my last Email has solved the error about missing extension.
> Now the following errors remain:
>
> [Jul  2 02:29:18] WARNING[18226][C-0001]: res_config_odbc.c:117
>> custom_prepare: SQL Prepare failed![SELECT * FROM sipusers WHERE host = ?
>> AND callbackextension = ? AND port = ?]
>> [Jul  2 02:29:22] WARNING[18269][C-]: app.c:1633
>> __ast_play_and_record: No audio available on SIP/domain.com-0000??
>
>
>
> On 2 July 2016 at 02:23, Nabeel  wrote:
>
>> The tutorial contains a mistake where the priority ordering in
>> extensions.conf should start with 1, not n:
>>
>> ; Voicemail
>>> exten => _VMR_.,1,Ringing
>>> exten => _VMR_.,n,Wait(1)
>>> exten => _VMR_.,n,Answer
>>> exten => _VMR_.,n,Wait(1)
>>> exten => _VMR_.,n,Voicemail(${EXTEN:4}|u)
>>> exten => _VMR_.,n,Hangup
>>
>> ; Allow users to call their Voicemail directly
>>> exten => VM_pickup,1,Ringing
>>> exten => VM_pickup,n,wait(1)
>>> exten => VM_pickup,n,VoicemailMain(${CALLERIDNUM}|s)
>>> exten => VM_pickup,n,Hangup
>>
>>
>>
>>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

The issue in my last Email has solved the error about missing extension.
Now the following errors remain:

[Jul  2 02:29:18] WARNING[18226][C-0001]: res_config_odbc.c:117
> custom_prepare: SQL Prepare failed![SELECT * FROM sipusers WHERE host = ?
> AND callbackextension = ? AND port = ?]
> [Jul  2 02:29:22] WARNING[18269][C-]: app.c:1633
> __ast_play_and_record: No audio available on SIP/domain.com-??



On 2 July 2016 at 02:23, Nabeel  wrote:

> The tutorial contains a mistake where the priority ordering in
> extensions.conf should start with 1, not n:
>
> ; Voicemail
>> exten => _VMR_.,1,Ringing
>> exten => _VMR_.,n,Wait(1)
>> exten => _VMR_.,n,Answer
>> exten => _VMR_.,n,Wait(1)
>> exten => _VMR_.,n,Voicemail(${EXTEN:4}|u)
>> exten => _VMR_.,n,Hangup
>
> ; Allow users to call their Voicemail directly
>> exten => VM_pickup,1,Ringing
>> exten => VM_pickup,n,wait(1)
>> exten => VM_pickup,n,VoicemailMain(${CALLERIDNUM}|s)
>> exten => VM_pickup,n,Hangup
>
>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

The tutorial contains a mistake where the priority ordering in
extensions.conf should start with 1, not n:

; Voicemail
> exten => _VMR_.,1,Ringing
> exten => _VMR_.,n,Wait(1)
> exten => _VMR_.,n,Answer
> exten => _VMR_.,n,Wait(1)
> exten => _VMR_.,n,Voicemail(${EXTEN:4}|u)
> exten => _VMR_.,n,Hangup

; Allow users to call their Voicemail directly
> exten => VM_pickup,1,Ringing
> exten => VM_pickup,n,wait(1)
> exten => VM_pickup,n,VoicemailMain(${CALLERIDNUM}|s)
> exten => VM_pickup,n,Hangup
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

Hi,

Adding 'limit 1' or 'limit 5' to the supusers mysql view resolves part of
the error, but I don't understand why that is and whether this is correct
for the setup. Maybe something to do with connection pooling?

Now the following errors remain:

[Jun 30 01:07:53] NOTICE[17067][C-] chan_sip.c: Call from '
> +447867958678' (162.248.6.216:12351 <http://162.249.6.206:12221/>) to
> extension 'VMR_+447479189410' rejected because extension not found in
> context 'default'.
> [Jun 30 01:07:53] WARNING[17112] res_config_odbc.c: SQL Prepare
> failed![SELECT * FROM sipusers WHERE name = ? AND host = ?]
> [Jun 30 01:07:53] WARNING[17112] res_odbc.c: Connection is down attempting
> to reconnect.


I am using OpenSIPS 2.2, not 2.3 as stated earlier.

Nabeel

On 30 June 2016 at 10:18, Bogdan-Andrei Iancu  wrote:

> Hi Nabeel,
>
> The "sipusers" mysql view (as per
> http://www.opensips.org/Documentation/Tutorials-OpenSIPSAsteriskIntegration-1-8#toc7
> ) has both the name and host fields - not sure why that query may fail.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 30.06.2016 07:19, Nabeel wrote:
>
> Hi Bogdan,
>
> I was able to install the latest versions of Asterisk (13.1) and Opensips
> (2.3) according to the tutorial, but when attempting to leave a voicemail I
> get the following errors:
>
>
>> [Jun 30 01:07:53] NOTICE[17067][C-] chan_sip.c: Call from '
>> +447867958678' (162.249.6.206:12221) to extension 'VMR_+447479189410'
>> rejected because extension not found in context 'default'.
>> [Jun 30 01:07:53] WARNING[17112] res_odbc.c: SetConnectAttr (Txn
>> isolation) returned an error: HY000: [MySQL][ODBC 5.2(w) Driver]You have an
>> error in your SQL syntax; check the manual that corresponds to your MariaDB
>> server version for the right syntax to use near '7' at line 1
>> [Jun 30 01:07:53] WARNING[17112] res_config_odbc.c: SQL Prepare
>> failed![SELECT * FROM sipusers WHERE name = ? AND host = ?]
>> [Jun 30 01:07:53] WARNING[17112] res_odbc.c: Connection is down
>> attempting to reconnect...
>>
>
>
> Also I had to change 'nat=yes' to 'nat=force_rport,comedia' as it is
> deprecated.
>
> Nabeel
>
>
> On 14 June 2016 at 11:08, Bogdan-Andrei Iancu < 
> bog...@opensips.org> wrote:
>
>> Hi Nabeel,
>>
>> We will update the tutorial for 2.2, but it should still match. Give it a
>> try and if you hit issues, just let me know.
>>
>> Regards,
>>
>> Bogdan-Andrei Iancu
>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>
>> On 12.06.2016 10:18, Nabeel wrote:
>>
>> Hi,
>>
>> I will be following this tutorial to integrate OpenSIPS and Asterisk:
>>
>>
>> http://www.opensips.org/Documentation/Tutorials-OpenSIPSAsteriskIntegration-1-8
>>
>> The tutorial mentions the use of OpenSIPS version 1.8 and Asterisk
>> version 1.8. I would like to know if I can use the latest versions of
>> OpenSIPS and Asterisk instead? Have there been changes to database
>> structure which can cause problems?
>>
>> Nabeel
>>
>>
>> ___
>> Users mailing 
>> listUsers@lists.opensips.orghttp://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] OpenSIPS and Asterisk integration versions

[Nabeel]

Hi Bogdan,

I was able to install the latest versions of Asterisk (13.1) and Opensips
(2.3) according to the tutorial, but when attempting to leave a voicemail I
get the following errors:


> [Jun 30 01:07:53] NOTICE[17067][C-] chan_sip.c: Call from
> '+447867958678' (162.249.6.206:12221) to extension 'VMR_+447479189410'
> rejected because extension not found in context 'default'.
> [Jun 30 01:07:53] WARNING[17112] res_odbc.c: SetConnectAttr (Txn
> isolation) returned an error: HY000: [MySQL][ODBC 5.2(w) Driver]You have an
> error in your SQL syntax; check the manual that corresponds to your MariaDB
> server version for the right syntax to use near '7' at line 1
> [Jun 30 01:07:53] WARNING[17112] res_config_odbc.c: SQL Prepare
> failed![SELECT * FROM sipusers WHERE name = ? AND host = ?]
> [Jun 30 01:07:53] WARNING[17112] res_odbc.c: Connection is down attempting
> to reconnect...
>


Also I had to change 'nat=yes' to 'nat=force_rport,comedia' as it is
deprecated.

Nabeel


On 14 June 2016 at 11:08, Bogdan-Andrei Iancu  wrote:

> Hi Nabeel,
>
> We will update the tutorial for 2.2, but it should still match. Give it a
> try and if you hit issues, just let me know.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 12.06.2016 10:18, Nabeel wrote:
>
> Hi,
>
> I will be following this tutorial to integrate OpenSIPS and Asterisk:
>
>
> http://www.opensips.org/Documentation/Tutorials-OpenSIPSAsteriskIntegration-1-8
>
> The tutorial mentions the use of OpenSIPS version 1.8 and Asterisk version
> 1.8. I would like to know if I can use the latest versions of OpenSIPS and
> Asterisk instead? Have there been changes to database structure which can
> cause problems?
>
> Nabeel
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] OpenSIPS and Asterisk integration versions

[Nabeel]

Hi,

I will be following this tutorial to integrate OpenSIPS and Asterisk:

http://www.opensips.org/Documentation/Tutorials-OpenSIPSAsteriskIntegration-1-8

The tutorial mentions the use of OpenSIPS version 1.8 and Asterisk version
1.8. I would like to know if I can use the latest versions of OpenSIPS and
Asterisk instead? Have there been changes to database structure which can
cause problems?

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

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

In that case, the answer to your question seems to be that the UDP packets
did not reach the OpenSIPS server, because nothing was added to the
OpenSIPS logs using debug level 4. All of this seems to point to the cause
being UDP packet fragmentation. Is this correct?
On 17 May 2016 4:24 pm, "Bogdan-Andrei Iancu"  wrote:

> The TCP/IP stack of your server may decide to drop an UDP packet if it
> cannot re-assemble it correctly (like not all the IP fragments were
> received).
> In such a case, you see the IP packets (carrying the fragments) on network
> level, but they are never delivered at application level.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 17.05.2016 16:05, Nabeel wrote:
>
> The next question - is this INVITE reaching your opensips script ? to be
>> sure that the OS delivers the UDP packet to the opensips application.
>
>
> I don't have any firewall on my server. Why would the UDP packet get
> blocked between entering the server and reaching opensips script? The
> opensips server is running without errors. Other calls work fine.
>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

>
> The next question - is this INVITE reaching your opensips script ? to be
> sure that the OS delivers the UDP packet to the opensips application.


I don't have any firewall on my server. Why would the UDP packet get
blocked between entering the server and reaching opensips script? The
opensips server is running without errors. Other calls work fine.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Do I have to rebuild my OpenSIPS installation completely just for this
module?
On 13 May 2016 12:16 am, "Ionut Ionita"  wrote:

> Hi,
>
> Did you build the module? Is the module loaded into your script?
>
> Regards,
>
> Ionut Ionita
> OpenSIPS Developer
>
> On 05/13/2016 02:09 AM, Nabeel wrote:
>
> Hi,
>
> I tried to use mc_compact but got this error:
>
> CRITICAL:core:yyerror: parse error in config file
> /usr/local//etc/opensips/opensips.cfg, line 201, column 12-13: unknown
> command , missing loadmodule?
>
> Then trying to load the module I got this error:
>
> CRITICAL:core:yyerror: parse error in config file
> /usr/local//etc/opensips/opensips.cfg, line 198, column 13-14: failed to
> load module mc_compact.so
>
> Also tried to load mc.so and I have zlib-devel dependency installed.
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
> ___
> 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] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Hi,

I tried to use mc_compact but got this error:

CRITICAL:core:yyerror: parse error in config file
/usr/local//etc/opensips/opensips.cfg, line 201, column 12-13: unknown
command , missing loadmodule?

Then trying to load the module I got this error:

CRITICAL:core:yyerror: parse error in config file
/usr/local//etc/opensips/opensips.cfg, line 198, column 13-14: failed to
load module mc_compact.so

Also tried to load mc.so and I have zlib-devel dependency installed.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Can packet fragmentation be verified (to be sure that it is packet
fragmentation)?
On 6 May 2016 5:28 pm, "Nabeel"  wrote:

> The trace I posted earlier is what I see with tcpdump when attempting a
> call. There is no other INVITE shown in the trace:
> http://pastebin.com/raw/C4iymTbh
>
> The trace seems to end abruptly in the middle of the SDP, so I think it
> could be due to packet fragmentation.
> On 6 May 2016 4:18 pm, "Bogdan-Andrei Iancu"  wrote:
>
>> So that meas the INVITE never gets to the callee ?? maybe it is not
>> properly routed .
>>
>> Do you see (with ngrep or tcpdump) the INVITE being sent out by opensips
>> towards callee ?
>>
>> Regards,
>>
>> Bogdan-Andrei Iancu
>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>
>> On 06.05.2016 12:56, Nabeel wrote:
>>
>> Hi,
>>
>> Thanks for the idea about packet compression. By 'call fails to connect',
>> I meant the call does not connect to the callee, ie. the callee's phone
>> does not ring after the INVITE (despite using TURN server).
>>
>> This was a public WiFi network and that was all I could get at the time.
>> I am using OpenSIPS version 2.1.
>>
>> Nabeel
>> On 6 May 2016 9:16 am, "Bogdan-Andrei Iancu"  wrote:
>>
>>> Hi,
>>>
>>> Hard to analyze a call based on the INVITE packet only :). Still the SIP
>>> signaling does not show any ALG interference (also not sure if the capture
>>> was done before or after the ALG). Also, what you mean by "call fails" ?no
>>> reply, negative reply , no audio ?
>>>
>>> Regards,
>>>
>>> Bogdan-Andrei Iancu
>>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>>
>>> On 05.05.2016 22:35, Nabeel wrote:
>>>
>>>
>>> Please check the following SIP trace taken within a WiFi network. The
>>> call fails to connect despite the INVITE request and using a non-standard
>>> port. Could this be caused by SIP ALG, or some unopened RTP port on the
>>> router?
>>>
>>> http://pastebin.com/raw/C4iymTbh
>>>
>>>
>>> ___
>>> Users mailing 
>>> listUsers@lists.opensips.orghttp://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] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

The trace I posted earlier is what I see with tcpdump when attempting a
call. There is no other INVITE shown in the trace:
http://pastebin.com/raw/C4iymTbh

The trace seems to end abruptly in the middle of the SDP, so I think it
could be due to packet fragmentation.
On 6 May 2016 4:18 pm, "Bogdan-Andrei Iancu"  wrote:

> So that meas the INVITE never gets to the callee ?? maybe it is not
> properly routed .
>
> Do you see (with ngrep or tcpdump) the INVITE being sent out by opensips
> towards callee ?
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 06.05.2016 12:56, Nabeel wrote:
>
> Hi,
>
> Thanks for the idea about packet compression. By 'call fails to connect',
> I meant the call does not connect to the callee, ie. the callee's phone
> does not ring after the INVITE (despite using TURN server).
>
> This was a public WiFi network and that was all I could get at the time. I
> am using OpenSIPS version 2.1.
>
> Nabeel
> On 6 May 2016 9:16 am, "Bogdan-Andrei Iancu"  wrote:
>
>> Hi,
>>
>> Hard to analyze a call based on the INVITE packet only :). Still the SIP
>> signaling does not show any ALG interference (also not sure if the capture
>> was done before or after the ALG). Also, what you mean by "call fails" ?no
>> reply, negative reply , no audio ?
>>
>> Regards,
>>
>> Bogdan-Andrei Iancu
>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>
>> On 05.05.2016 22:35, Nabeel wrote:
>>
>>
>> Please check the following SIP trace taken within a WiFi network. The
>> call fails to connect despite the INVITE request and using a non-standard
>> port. Could this be caused by SIP ALG, or some unopened RTP port on the
>> router?
>>
>> http://pastebin.com/raw/C4iymTbh
>>
>>
>> ___
>> Users mailing 
>> listUsers@lists.opensips.orghttp://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] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Hi,

Thanks for the idea about packet compression. By 'call fails to connect', I
meant the call does not connect to the callee, ie. the callee's phone does
not ring after the INVITE (despite using TURN server).

This was a public WiFi network and that was all I could get at the time. I
am using OpenSIPS version 2.1.

Nabeel
On 6 May 2016 9:16 am, "Bogdan-Andrei Iancu"  wrote:

> Hi,
>
> Hard to analyze a call based on the INVITE packet only :). Still the SIP
> signaling does not show any ALG interference (also not sure if the capture
> was done before or after the ALG). Also, what you mean by "call fails" ?no
> reply, negative reply , no audio ?
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 05.05.2016 22:35, Nabeel wrote:
>
>
> Please check the following SIP trace taken within a WiFi network. The call
> fails to connect despite the INVITE request and using a non-standard port.
> Could this be caused by SIP ALG, or some unopened RTP port on the router?
>
> http://pastebin.com/raw/C4iymTbh
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://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] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Yes, it is definitely listening on the non-standard port:

# netstat -lnp | grep opensips
udp0  0 162.xxx.x.110:12341 0.0.0.0:* 1750/opensips

In fact, it even seems to register correctly, but the INVITE does not get
through.

I have OpenSIPS set on debug level 4 but did not see anything in the log.


On 5 May 2016 at 21:14, Tito Cumpen  wrote:

> Nabeel,
>
>
> Did you verify that your opensips server is listening on this non Standard
> port ?
>
> run
>
> netstat -lnp | grep opensips
>
> On Thu, May 5, 2016 at 4:12 PM, Nabeel  wrote:
>
>> Tito Campen, I took the trace on my OpenSIPS server, which is the
>> receiving proxy. However, OpenSIPS did not add anything to the server's log
>> itself.
>>
>> On 5 May 2016 at 21:08, Tito Cumpen  wrote:
>>
>>> Nabel, You should should take a trace at the receiving proxy to verify
>>> the traffic is even getting there. If there is no sdp received from the UAS
>>> you would not see rtp traversing at all. Using non Standard points doesn't
>>> assure you that messaging traffic will traverse.
>>>
>>> On Thu, May 5, 2016 at 3:54 PM, Tito Cumpen  wrote:
>>>
>>>> Adding to Bogdan's point I am successfully using sip tls on port 443
>>>> without any issues as of yet. It's bypassing some  isp enforced algs as
>>>> well as those enforced by local routers. :-).
>>>>
>>>>
>>>>
>>>>
>>>> On Thu, May 5, 2016 at 3:35 PM, Nabeel  wrote:
>>>>
>>>>>
>>>>> Please check the following SIP trace taken within a WiFi network. The
>>>>> call fails to connect despite the INVITE request and using a non-standard
>>>>> port. Could this be caused by SIP ALG, or some unopened RTP port on the
>>>>> router?
>>>>>
>>>>> http://pastebin.com/raw/C4iymTbh
>>>>>
>>>>> ___
>>>>> 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
>>>
>>>
>>
>> ___
>> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Tito Campen, I took the trace on my OpenSIPS server, which is the receiving
proxy. However, OpenSIPS did not add anything to the server's log itself.

On 5 May 2016 at 21:08, Tito Cumpen  wrote:

> Nabel, You should should take a trace at the receiving proxy to verify the
> traffic is even getting there. If there is no sdp received from the UAS you
> would not see rtp traversing at all. Using non Standard points doesn't
> assure you that messaging traffic will traverse.
>
> On Thu, May 5, 2016 at 3:54 PM, Tito Cumpen  wrote:
>
>> Adding to Bogdan's point I am successfully using sip tls on port 443
>> without any issues as of yet. It's bypassing some  isp enforced algs as
>> well as those enforced by local routers. :-).
>>
>>
>>
>>
>> On Thu, May 5, 2016 at 3:35 PM, Nabeel  wrote:
>>
>>>
>>> Please check the following SIP trace taken within a WiFi network. The
>>> call fails to connect despite the INVITE request and using a non-standard
>>> port. Could this be caused by SIP ALG, or some unopened RTP port on the
>>> router?
>>>
>>> http://pastebin.com/raw/C4iymTbh
>>>
>>> ___
>>> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Please read why UDP is better for VoIP:
https://www.onsip.com/blog/udp-versus-tcp-for-voip
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Please check the following SIP trace taken within a WiFi network. The call
fails to connect despite the INVITE request and using a non-standard port.
Could this be caused by SIP ALG, or some unopened RTP port on the router?

http://pastebin.com/raw/C4iymTbh
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

A possible solution to this seems to be a 'SIP tunnel' server. The server
would tunnel the SIP and UDP packets over a common TCP port such as 80 or
443, which are more likely to be open and unblocked on Wi-Fi routers for
browsing, Email, etc. The tunnel server would then send this data to
OpenSIPS over UDP. There is a SIP tunnel server for Windows here:
http://siptunnel.sourceforge.net/, however I was not able to find a similar
tunnel server for Linux? Does OpenSIPS have any functions relating to this?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Thanks for the suggestions of using TLS or changing the port. I changed the
port, but some routers are still able to mess with the SIP headers. I would
have used TLS, if not for two reasons:

1. ICE protocol was originally designed for UDP according to RFC5245, and
it seems to work better with UDP.

2. The SIP servers I have used (OpenSIPS and Repro) seem to be more stable
with UDP compared to TLS (they do not randomly drop connections, throw
unusual errors in the logs, etc.)

I may try TLS again, but it would be better if there is an alternative
workaround for UDP.

On 2 May 2016 at 13:33, Patrick Wakano  wrote:

> Using TLS!
> Also configuring your systems/devices to use other port than 5060 may do
> the trick...
>
> On Mon, May 2, 2016 at 9:14 AM, Nabeel  wrote:
>
>> Hi,
>>
>> Other than using rtpproxy/NAThelper modules, is there any way to
>> bypass/workaround SIP ALG enabled on many WiFi routers? Although SIP ALG
>> was designed to help with NAT, in most cases it does the opposite and
>> breaks SIP.
>>
>> Nabeel
>>
>> ___
>> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] How to overcome SIP ALG on Wi-Fi routers

[Nabeel]

Hi,

Other than using rtpproxy/NAThelper modules, is there any way to
bypass/workaround SIP ALG enabled on many WiFi routers? Although SIP ALG
was designed to help with NAT, in most cases it does the opposite and
breaks SIP.

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

Re: [OpenSIPS-Users] rtpproxy TCP connection

[Nabeel]

Could this also explain why other TURN servers don't work so well over TCP
with OpenSIPS?


On 18 February 2016 at 12:09, Răzvan Crainea  wrote:

> Hi Gomtesh!
>
> Currently only UDP and UNIX datagrams sockets are supported.
>
> Best regards,
> Răzvan
>
>
> On 02/18/2016 01:15 PM, Gomtesh Jain wrote:
>
> Is it possible to make tcp connection to rtpproxy from opensips ? I am
> trying
>  modparam("rtpproxy", "rtpproxy_sock", "tcp:127.0.0.1:2")
>
> But it is not working .
> any suggestions ?
>
> Thanks,
> Gomtesh
>
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
> --
> Răzvan Crainea
> OpenSIPS Core Developerhttp://www.opensips-solutions.com
>
>
> ___
> 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] How to TLS ?

[Nabeel]

Hi,

That option is only required if you want to enable "Mutual (two-way) client
authentication' and is not normally necessary when using TLS. Most of these
clients don't seem to support two way authentication. You can have this
option disabled:
modparam("proto_tls","require_cert", "0").

477 error in my experience is usually a temporary connection error related
to  TLS, but not directly related to configuration.

Nabeel
On 12 Feb 2016 6:45 am, "Hamid Hashmi"  wrote:

> Nabeel
>
> I dont know how to present a certificate from client. I have tried using
> Xoiper (Android - Free), SFLphone (Ubuntu) and CsipSimple (Android) but
> there was no options set a public key.
>
> Now I am using CA signed certificates in opensips with disabled flags of
> verify_cert and require_cert, having an error of *477 Send failed
> (477/TM). *
>
> *Hamid R. Hashmi*
> Software Engineer - VoIP
> Vopium A/S
>
>
> --
> Date: Tue, 9 Feb 2016 08:48:41 +
> From: nabeelshik...@gmail.com
> To: users@lists.opensips.org
> Subject: Re: [OpenSIPS-Users] How to TLS ?
>
> Hi,
>
> Does the client present a client certificate? If not, then with
> modparam("proto_tls","require_cert", "1"), OpenSIPS misleadingly logs:
> 'failed to accept: rejected by client'.  What it actually means is that
> the client failed to present a certificate.
> On 9 Feb 2016 6:06 am, "Hamid Hashmi"  wrote:
>
> It will be a great help if you please help me in configuring TLS. I have
> followed this <http://www.opensips.org/Documentation/Tutorials-TLS-2-1>
> to configure TLS but could not able to verify certificates.
>
> its working if disable following flags
>
> modparam("proto_tls","verify_cert", "0")
> modparam("proto_tls","require_cert", "0")
>
> BUT not verifying certificates. Please see logs
> <http://pastebin.com/qmXZjSy2> if enabled
>
> modparam("proto_tls","verify_cert", "1")
> modparam("proto_tls","require_cert", "1")
>
> then have following ERROR
>
> Feb  9 05:57:14 comoyo-dev-ec2-siplb SIPLB[29867]: 
> [udp:keepalive@192.168.26.181:8000 <http://192.168.26.181:8000>]: Receive 
> request OPTIONS from local server [192.168.26.181]
> Feb  9 05:57:14 comoyo-dev-ec2-siplb SIPLB[29868]: 
> ERROR:proto_tls:tls_accept: New TLS connection from 115.186.93.1:47015 failed 
> to accept: rejected by client
> Feb  9 05:57:14 comoyo-dev-ec2-siplb SIPLB[29868]: 
> ERROR:proto_tls:tls_read_req: failed to do pre-tls reading
> Feb  9 05:57:17 comoyo-dev-ec2-siplb SIPLB[29863]: 
> [tcp:siplb@192.168.26.180:6080 <http://192.168.26.180:6080>]: In LOCAL Route 
> sending OPTIONS to 192.168.26.181
> Feb  9 05:57:17 comoyo-dev-ec2-siplb SIPLB[29863]: 
> INFO:core:probe_max_sock_buff: using snd buffer of 244 kb
> Feb  9 05:57:17 comoyo-dev-ec2-siplb SIPLB[29863]: 
> INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 17
>
> Regards
> *Hamid R. Hashmi*
>
>
> ___
> 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
>
> ___
> 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] How to TLS ?

[Nabeel]

Hi,

Does the client present a client certificate? If not, then with
modparam("proto_tls","require_cert", "1"), OpenSIPS misleadingly logs:
'failed to accept: rejected by client'.  What it actually means is that the
client failed to present a certificate.
On 9 Feb 2016 6:06 am, "Hamid Hashmi"  wrote:

> It will be a great help if you please help me in configuring TLS. I have
> followed this 
> to configure TLS but could not able to verify certificates.
>
> its working if disable following flags
>
> modparam("proto_tls","verify_cert", "0")
> modparam("proto_tls","require_cert", "0")
>
> BUT not verifying certificates. Please see logs
>  if enabled
>
> modparam("proto_tls","verify_cert", "1")
> modparam("proto_tls","require_cert", "1")
>
> then have following ERROR
>
> Feb  9 05:57:14 comoyo-dev-ec2-siplb SIPLB[29867]: 
> [udp:keepalive@192.168.26.181:8000]: Receive request OPTIONS from local 
> server [192.168.26.181]
> Feb  9 05:57:14 comoyo-dev-ec2-siplb SIPLB[29868]: 
> ERROR:proto_tls:tls_accept: New TLS connection from 115.186.93.1:47015 failed 
> to accept: rejected by client
> Feb  9 05:57:14 comoyo-dev-ec2-siplb SIPLB[29868]: 
> ERROR:proto_tls:tls_read_req: failed to do pre-tls reading
> Feb  9 05:57:17 comoyo-dev-ec2-siplb SIPLB[29863]: 
> [tcp:siplb@192.168.26.180:6080]: In LOCAL Route sending OPTIONS to 
> 192.168.26.181
> Feb  9 05:57:17 comoyo-dev-ec2-siplb SIPLB[29863]: 
> INFO:core:probe_max_sock_buff: using snd buffer of 244 kb
> Feb  9 05:57:17 comoyo-dev-ec2-siplb SIPLB[29863]: 
> INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 17
>
> Regards
> *Hamid R. Hashmi*
>
>
> ___
> 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] Linphone and OpenSIPS over TCP

[Nabeel]

Hi,

I understand that OpenSIPS supports Mediaproxy which only relays RTP
packets over UDP.  My concern now is that due to Mediaproxy only supporting
UDP, OpenSIPS might have this limitation built into it for all TURN
servers, including Coturn.

In Linphone settings, there is a 'Stun server' option, and although it does
not state 'TURN server' here, I found that in most clients which do support
a TURN server, they still tend to refer to this as a 'Stun server' in their
settings.
On 6 Feb 2016 8:56 am, "Adrian Georgescu"  wrote:

> OpenSIPS does not need to interact in anyway with a TURN server. A TURN
> server is used by the SIP client that has built-in TURN functionality.
> While they may run side by side, the only interaction of a sip proxy and
> TURN server is sharing the same database with credentials in order to
> authenticate the end-users.
>
> What MediaProxy does, it simulates a TURN server candidate by inserting it
> into an offer generated by a SIP client that has ICE support. But it is not
> real TURN in the sense that it does not implement the TURN protocol (which
> does a lot of other things, not just the relaying packets over UDP part),
> and the client is unaware of this TURN candidate insertion. Practically,
> MediaProxy only relays RTP packets over UDP and is fouling both end-points
> into believing that during the ICE negotiation there is a relay server that
> can be used when end-to-end RTP does not work. One cannot get RTP over TCP
> running with this hybrid model.
>
> If you want TCP, you need a real TURN client with a real TURN server.
>
> Regards,
> Adrian
>
>
> On 06 Feb 2016, at 01:07, Nabeel  wrote:
>
> On 3 February 2016 at 23:42, sevpal  wrote:
>
> Opensips interacts with the TURN in server MediaProxy only.
>>
>
>
> That's not completely true, because the TURN server works with OpenSIPS
> when using UDP for calls. It just doesn't work - or only partially works -
> with TCP/TLS.  I tested by changing configurations in the TURN server to
> use TCP relays/listeners only, but it still only works with UDP. This leads
> me to believe that it is a limitation in OpenSIPS that it only handles the
> TURN server properly when calls are using UDP, not TCP/TLS.
>
> Coturn is probably the most comprehensive TURN server available today. Is
> there any chance that OpenSIPS will fully support this in the future?
> ___
> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

On 3 February 2016 at 23:42, sevpal  wrote:

Opensips interacts with the TURN in server MediaProxy only.
>


That's not completely true, because the TURN server works with OpenSIPS
when using UDP for calls. It just doesn't work - or only partially works -
with TCP/TLS.  I tested by changing configurations in the TURN server to
use TCP relays/listeners only, but it still only works with UDP. This leads
me to believe that it is a limitation in OpenSIPS that it only handles the
TURN server properly when calls are using UDP, not TCP/TLS.

Coturn is probably the most comprehensive TURN server available today. Is
there any chance that OpenSIPS will fully support this in the future?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

In the SDP of the call, I see UDP relays being used:

a=candidate:1 1 UDP 2130706431 10.180.107.181 7076 typ host a=candidate:1 2
UDP 2130706430 10.180.107.181 7077 typ host a=candidate:2 1 UDP 1694498815
92.40.248.63 51755 typ srflx raddr 10.180.107.181

For calls over TCP and TLS, should the relays be using TCP instead of UDP?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

Hi,

Please see the tcpdump trace below.  Can this be caused by incorrect use of
the TURN server? I am convinced that the problem relates to use of the TURN
server, but not sure if the problem is caused by OpenSIPS or the TURN
server itself.

http://pastebin.com/HPZ7nRYS
The problem is not the 5 seconds timeout at sip level, is the fact that the
end point you are trying to connect to via TCP does not accept the
connection. You can check that via tcpdump.

Regards,

Bogdan-Andrei Iancu
OpenSIPS Founder and Developerhttp://www.opensips-solutions.com

On 02.02.2016 11:17, Nabeel wrote:

Hi Bogdan,

Even if I remove the 5 second timeout by removing #modparam("tm",
"fr_inv_timeout", 30), the timeout occurs after about 20 seconds. What do
you suggest is the solution?
Hi Nabeel,

Your OpenSIPS tries to connect via TCP to the destination and to avoid
blocking it is doing it async:


Feb  1 19:22:13 server1 opensips: Feb  1 19:22:13 [20587]
DBG:core:print_ip: tcpconn_new: new tcp connection to: 188.29.165.162
Feb  1 19:22:13 server1 opensips: Feb  1 19:22:13 [20587]
DBG:core:tcpconn_new: on port 58985, proto 2
Feb  1 19:22:13 server1 opensips: Feb  1 19:22:13 [20587]
DBG:core:proto_tcp_send: Successfully started async connection

After that, in 5 seconds, the final timer hits (as timeout for no reply),
while the TCP connect still haven;t finished (so there is no actual packet
sent out).

Regards,

Bogdan-Andrei Iancu
OpenSIPS Founder and Developerhttp://www.opensips-solutions.com

On 01.02.2016 21:35, Nabeel wrote:

Hi Bogdan,

Below is the requested log for TCP call attempt.  User  is trying to
call user  via OpenSIPS server 162.248.6.120 :

http://pastebin.com/UQ9mEemd

On 1 February 2016 at 09:41, Bogdan-Andrei Iancu 
wrote:

> Hi,
>
> I strongly suggest to look into the opensips logs and see what opensips
> try to do with the call. Based on your saying (that you see a timeout), I
> suspect your OpenSIPS tries to deliver the call over TCP to a destination
> which does not listen on TCP.
> If you do not know hoe to interpret the logs, run opensips in debug=4
> mode, upload the logs corresponding the INVITE execution and provide the
> link.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 31.01.2016 16:28, Nabeel wrote:
>
> Without using alias=domain.com, TCP still does not work.  My initial
> request for someone to test this using Linphone remains. Please test and
> let me know if you can call using TCP with OpenSIPS listening on an IP
> address.
>
> On 31 January 2016 at 09:28, Nabeel  wrote:
>
>> On further testing, using the IP address instead of the domain name in
>> the URI setting of Linphone works with TCP, so I think this might be to do
>> with SRV/NAPTR records associated with the domain.
>>
>> On 31 January 2016 at 08:29, Nabeel  wrote:
>>
>>> Hello,
>>>
>>> There seems to be a problem with calls over TCP using Linphone, and
>>> since Linphone is a popular open source application, I would like someone
>>> to please verify this problem. Calls work fine with Linphone over UDP, but
>>> after registering with TCP using the same credentials, calls do not connect
>>> at all and lead to a request timeout.  A request timeout does not say much
>>> about the cause, but in this case I suspect there is something wrong with
>>> TCP on the server side. I would like someone to please install Linphone on
>>> your phone and connect to your OpenSIPS server using UDP and TCP.  Please
>>> report here if the calls work over both transports.
>>>
>>
>>
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://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] Linphone and OpenSIPS over TCP

[Nabeel]

Hi Bogdan,

Even if I remove the 5 second timeout by removing #modparam("tm",
"fr_inv_timeout", 30), the timeout occurs after about 20 seconds. What do
you suggest is the solution?
Hi Nabeel,

Your OpenSIPS tries to connect via TCP to the destination and to avoid
blocking it is doing it async:


Feb  1 19:22:13 server1 opensips: Feb  1 19:22:13 [20587]
DBG:core:print_ip: tcpconn_new: new tcp connection to: 188.29.165.162
Feb  1 19:22:13 server1 opensips: Feb  1 19:22:13 [20587]
DBG:core:tcpconn_new: on port 58985, proto 2
Feb  1 19:22:13 server1 opensips: Feb  1 19:22:13 [20587]
DBG:core:proto_tcp_send: Successfully started async connection

After that, in 5 seconds, the final timer hits (as timeout for no reply),
while the TCP connect still haven;t finished (so there is no actual packet
sent out).

Regards,

Bogdan-Andrei Iancu
OpenSIPS Founder and Developerhttp://www.opensips-solutions.com

On 01.02.2016 21:35, Nabeel wrote:

Hi Bogdan,

Below is the requested log for TCP call attempt.  User  is trying to
call user  via OpenSIPS server 162.248.6.120 :

http://pastebin.com/UQ9mEemd

On 1 February 2016 at 09:41, Bogdan-Andrei Iancu 
wrote:

> Hi,
>
> I strongly suggest to look into the opensips logs and see what opensips
> try to do with the call. Based on your saying (that you see a timeout), I
> suspect your OpenSIPS tries to deliver the call over TCP to a destination
> which does not listen on TCP.
> If you do not know hoe to interpret the logs, run opensips in debug=4
> mode, upload the logs corresponding the INVITE execution and provide the
> link.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 31.01.2016 16:28, Nabeel wrote:
>
> Without using alias=domain.com, TCP still does not work.  My initial
> request for someone to test this using Linphone remains. Please test and
> let me know if you can call using TCP with OpenSIPS listening on an IP
> address.
>
> On 31 January 2016 at 09:28, Nabeel  wrote:
>
>> On further testing, using the IP address instead of the domain name in
>> the URI setting of Linphone works with TCP, so I think this might be to do
>> with SRV/NAPTR records associated with the domain.
>>
>> On 31 January 2016 at 08:29, Nabeel  wrote:
>>
>>> Hello,
>>>
>>> There seems to be a problem with calls over TCP using Linphone, and
>>> since Linphone is a popular open source application, I would like someone
>>> to please verify this problem. Calls work fine with Linphone over UDP, but
>>> after registering with TCP using the same credentials, calls do not connect
>>> at all and lead to a request timeout.  A request timeout does not say much
>>> about the cause, but in this case I suspect there is something wrong with
>>> TCP on the server side. I would like someone to please install Linphone on
>>> your phone and connect to your OpenSIPS server using UDP and TCP.  Please
>>> report here if the calls work over both transports.
>>>
>>
>>
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://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] Linphone and OpenSIPS over TCP

[Nabeel]

Hamid, to answer your question about contact, here are the contacts of
callee and caller from memory:


AOR:: 
> Contact:: 
> sip:@188.29.164.62:36612;app-id=622464153529;pn-type=google;pn-tok=APA91bF-OsPp0vEMG1m_RZLhIaLNY90KYav0So_8a8Lm0rm-fUecuikFF5Prmnb_m5t9E4Cfavzssp-jkrzw9Y-VZ1FjqUSVJaTUhnE5FtSiJhUqkpgSprE;transport=tcp
> Q=
> Expires:: 86392
> Callid:: k8R9KdUDVj
> Cseq:: 21
> User-agent:: LinphoneAndroid/2599 (belle-sip/1.4.2)
> Received:: sip:188.29.164.62:36612;transport=TCP
> State:: CS_NEW
> Flags:: 0
> Cflags:: NAT
> Socket:: tcp:162.248.6.120:5060
> Methods:: 4294967295
> SIP_instance::
> 



 AOR:: 
> Contact:: 
> sip:@92.40.248.63:58688;app-id=622464153529;pn-type=google;pn-tok=APA91bGt_p4GcPUNLZsI_4YyvV8DzDYisZfBr9tVu7WpbWI1KOJ3Y5EUa0S-h9Zp8uq76TL2zpLhm5ZOkoyJ3rUC63903lOq050xXRt4LMz-059gwfQWiNc;transport=tcp
> Q=
> Expires:: 86383
> Callid:: 1ts0tZS4vk
> Cseq:: 21
> User-agent:: LinphoneAndroid/2599 (belle-sip/1.4.2)
> Received:: sip:92.40.248.63:58688;transport=TCP
> State:: CS_NEW
> Flags:: 0
> Cflags:: NAT
> Socket:: tcp:162.248.6.120:5060
> Methods:: 4294967295
> SIP_instance::
> 
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

Hi Hamid,

I am not using the location table, I am using memory only to store the
contacts.

If you use TCP within the same network, where a TURN server is not required
(TURN server not set in Linphone settings), I found that the calls then
work depending on what the network allows. Within the same Wi-Fi network,
in my case audio calls work but not video calls. But using 3G/4G only, with
or without a TURN server set in Linphone settings, calls do not work at
all. This is why I think the behaviour of TURN server has something to do
with this. To those people stating that TCP works with Linphone, please
test over 3G/4G where a TURN server might be required.


On 2 Feb 2016 7:07 am, "Hamid Hashmi"  wrote:

> Nabeel I have been using Linphone 3.6.1 with opensips for a long time. And
> its working fine on both UDP and TCP. I have gone through your logs, there
> is log line DBG:tm:matching_3261: RFC3261 transaction matching failed
> Please check contact of your "to number" in location table.
>
>
> *Hamid R. Hashmi*
> Software Engineer - VoIP
> Vopium A/S
>
>
> --
> Date: Mon, 1 Feb 2016 19:35:53 +
> From: nabeelshik...@gmail.com
> To: bog...@opensips.org; users@lists.opensips.org
> Subject: Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP
>
> Hi Bogdan,
>
> Below is the requested log for TCP call attempt.  User  is trying to
> call user  via OpenSIPS server 162.248.6.120 :
>
> http://pastebin.com/UQ9mEemd
>
> On 1 February 2016 at 09:41, Bogdan-Andrei Iancu 
> wrote:
>
> Hi,
>
> I strongly suggest to look into the opensips logs and see what opensips
> try to do with the call. Based on your saying (that you see a timeout), I
> suspect your OpenSIPS tries to deliver the call over TCP to a destination
> which does not listen on TCP.
> If you do not know hoe to interpret the logs, run opensips in debug=4
> mode, upload the logs corresponding the INVITE execution and provide the
> link.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 31.01.2016 16:28, Nabeel wrote:
>
> Without using alias=domain.com, TCP still does not work.  My initial
> request for someone to test this using Linphone remains. Please test and
> let me know if you can call using TCP with OpenSIPS listening on an IP
> address.
>
> On 31 January 2016 at 09:28, Nabeel  wrote:
>
> On further testing, using the IP address instead of the domain name in the
> URI setting of Linphone works with TCP, so I think this might be to do with
> SRV/NAPTR records associated with the domain.
>
> On 31 January 2016 at 08:29, Nabeel < 
> nabeelshik...@gmail.com> wrote:
>
> Hello,
>
> There seems to be a problem with calls over TCP using Linphone, and since
> Linphone is a popular open source application, I would like someone to
> please verify this problem. Calls work fine with Linphone over UDP, but
> after registering with TCP using the same credentials, calls do not connect
> at all and lead to a request timeout.  A request timeout does not say much
> about the cause, but in this case I suspect there is something wrong with
> TCP on the server side. I would like someone to please install Linphone on
> your phone and connect to your OpenSIPS server using UDP and TCP.  Please
> report here if the calls work over both transports.
>
>
>
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
>
> ___ 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

Hi Bogdan,

Below is the requested log for TCP call attempt.  User  is trying to
call user  via OpenSIPS server 162.248.6.120 :

http://pastebin.com/UQ9mEemd

On 1 February 2016 at 09:41, Bogdan-Andrei Iancu 
wrote:

> Hi,
>
> I strongly suggest to look into the opensips logs and see what opensips
> try to do with the call. Based on your saying (that you see a timeout), I
> suspect your OpenSIPS tries to deliver the call over TCP to a destination
> which does not listen on TCP.
> If you do not know hoe to interpret the logs, run opensips in debug=4
> mode, upload the logs corresponding the INVITE execution and provide the
> link.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 31.01.2016 16:28, Nabeel wrote:
>
> Without using alias=domain.com, TCP still does not work.  My initial
> request for someone to test this using Linphone remains. Please test and
> let me know if you can call using TCP with OpenSIPS listening on an IP
> address.
>
> On 31 January 2016 at 09:28, Nabeel  wrote:
>
>> On further testing, using the IP address instead of the domain name in
>> the URI setting of Linphone works with TCP, so I think this might be to do
>> with SRV/NAPTR records associated with the domain.
>>
>> On 31 January 2016 at 08:29, Nabeel < 
>> nabeelshik...@gmail.com> wrote:
>>
>>> Hello,
>>>
>>> There seems to be a problem with calls over TCP using Linphone, and
>>> since Linphone is a popular open source application, I would like someone
>>> to please verify this problem. Calls work fine with Linphone over UDP, but
>>> after registering with TCP using the same credentials, calls do not connect
>>> at all and lead to a request timeout.  A request timeout does not say much
>>> about the cause, but in this case I suspect there is something wrong with
>>> TCP on the server side. I would like someone to please install Linphone on
>>> your phone and connect to your OpenSIPS server using UDP and TCP.  Please
>>> report here if the calls work over both transports.
>>>
>>
>>
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://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] Linphone and OpenSIPS over TCP

[Nabeel]

sevpal,

Thanks for the suggestions. I didn't have the domain in the domain table of
the database. However, I have added this and the problem is not solved.
The calls otherwise work without it, so I'm not sure why the domain table
is necessary.

I'm still working on the theory that the TURN server behaves differently
with TCP, UDP and TLS.  It is very strange behaviour, but Linphone
definitely shows this discrepancy with TCP using the Coturn TURN server.

On 31 January 2016 at 17:11, sevpal  wrote:

> It’s just a configuration issue you are having, start by:
>
> 1. In windows, do an “nslookup your.domain.net”
> if the returned IP’s are not what you expect, then correct this. You
> may also want to do a reverse lookup “nslookup xxx.xxx.xxx.xxx” to return
> its domain name.
>
> 2. Configure your Opensips to listen on these IP’s
>
> 3. Add “your.domain.net” in your domain table (these are the domains your
> sips is responsible for, IP’s can go in this table but not recommended if
> you are strictly using domain names in the clients for authentication)
>
> By the way, TCP works with Linphone very well.
>
>
> *From:* Nabeel 
> *Sent:* Sunday, January 31, 2016 9:28 AM
> *To:* OpenSIPS users mailling list 
> *Subject:* Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP
>
> Without using alias=domain.com, TCP still does not work.  My initial
> request for someone to test this using Linphone remains. Please test and
> let me know if you can call using TCP with OpenSIPS listening on an IP
> address.
>
> On 31 January 2016 at 09:28, Nabeel  wrote:
>
>> On further testing, using the IP address instead of the domain name in
>> the URI setting of Linphone works with TCP, so I think this might be to do
>> with SRV/NAPTR records associated with the domain.
>>
>> On 31 January 2016 at 08:29, Nabeel  wrote:
>>
>>> Hello,
>>>
>>> There seems to be a problem with calls over TCP using Linphone, and
>>> since Linphone is a popular open source application, I would like someone
>>> to please verify this problem. Calls work fine with Linphone over UDP, but
>>> after registering with TCP using the same credentials, calls do not connect
>>> at all and lead to a request timeout.  A request timeout does not say much
>>> about the cause, but in this case I suspect there is something wrong with
>>> TCP on the server side. I would like someone to please install Linphone on
>>> your phone and connect to your OpenSIPS server using UDP and TCP.  Please
>>> report here if the calls work over both transports.
>>>
>>
>>
>
>
> --
> ___
> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

I now believe this may be a discrepancy in the way my TURN server (Coturn)
is used over different protocols, as follows:

Using UDP, all calls use Coturn correctly and can establish calls without
problems.

Using TLS, calls within 15 minutes of clients' registration use Coturn
correctly and the calls can connect, but after 15 minutes of registration,
Coturn cannot establish the calls.

Using TCP, Coturn cannot establish the calls at all.

Please let me know if these are known issues with OpenSIPS+TURN server, and
if anything can be done on the OpenSIPS side to fix this.  I have contacted
the Coturn forum regarding this.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

Without using alias=domain.com, TCP still does not work.  My initial
request for someone to test this using Linphone remains. Please test and
let me know if you can call using TCP with OpenSIPS listening on an IP
address.

On 31 January 2016 at 09:28, Nabeel  wrote:

> On further testing, using the IP address instead of the domain name in the
> URI setting of Linphone works with TCP, so I think this might be to do with
> SRV/NAPTR records associated with the domain.
>
> On 31 January 2016 at 08:29, Nabeel  wrote:
>
>> Hello,
>>
>> There seems to be a problem with calls over TCP using Linphone, and since
>> Linphone is a popular open source application, I would like someone to
>> please verify this problem. Calls work fine with Linphone over UDP, but
>> after registering with TCP using the same credentials, calls do not connect
>> at all and lead to a request timeout.  A request timeout does not say much
>> about the cause, but in this case I suspect there is something wrong with
>> TCP on the server side. I would like someone to please install Linphone on
>> your phone and connect to your OpenSIPS server using UDP and TCP.  Please
>> report here if the calls work over both transports.
>>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

On further testing, using the IP address instead of the domain name in the
URI setting of Linphone works with TCP, so I think this might be to do with
SRV/NAPTR records associated with the domain.

On 31 January 2016 at 08:29, Nabeel  wrote:

> Hello,
>
> There seems to be a problem with calls over TCP using Linphone, and since
> Linphone is a popular open source application, I would like someone to
> please verify this problem. Calls work fine with Linphone over UDP, but
> after registering with TCP using the same credentials, calls do not connect
> at all and lead to a request timeout.  A request timeout does not say much
> about the cause, but in this case I suspect there is something wrong with
> TCP on the server side. I would like someone to please install Linphone on
> your phone and connect to your OpenSIPS server using UDP and TCP.  Please
> report here if the calls work over both transports.
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] Linphone and OpenSIPS over TCP

[Nabeel]

Hello,

There seems to be a problem with calls over TCP using Linphone, and since
Linphone is a popular open source application, I would like someone to
please verify this problem. Calls work fine with Linphone over UDP, but
after registering with TCP using the same credentials, calls do not connect
at all and lead to a request timeout.  A request timeout does not say much
about the cause, but in this case I suspect there is something wrong with
TCP on the server side. I would like someone to please install Linphone on
your phone and connect to your OpenSIPS server using UDP and TCP.  Please
report here if the calls work over both transports.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Request Timeout on INVITE After Idle Registration

[Nabeel]

and here is the more detailed SIP trace for UDP:

http://pastebin.com/UfQJJz3Y



On 13 January 2016 at 06:19, Nabeel  wrote:

> Hi Bogdan,
>
> I changed log_stderror=yes and log_facility=LOG_DAEMON.  Now I see some
> more in the log.  Do you see anything obviously wrong?
>
> http://pastebin.com/MzJW1P1S
>
> On 12 January 2016 at 09:10, Bogdan-Andrei Iancu 
> wrote:
>
>> Hi Nabeel,
>>
>> Be sure you are looking into the right log file - maybe the debug level
>> is redirected by your syslog to another log file... Debug level 4 is the
>> most verbose one in opensips.
>>
>> Regards,
>>
>> Bogdan-Andrei Iancu
>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>
>> On 08.01.2016 21:04, Nabeel wrote:
>>
>> Hi Bogdan,
>>
>> I have the following near the top of my config file:
>>
>> ## Global Parameters #
>>
>> debug=4
>> log_stderror=no
>> log_facility=LOG_LOCAL1
>>
>> The log I posted earlier is from opensips running with these
>> configurations.
>> On 8 Jan 2016 3:49 pm, "Bogdan-Andrei Iancu"  wrote:
>>
>>> Hi Nabeel,
>>>
>>> have you tried running opensips is debug mode (level 4) to see what it
>>> is doing with the request ?
>>>
>>> Regards,
>>>
>>> Bogdan-Andrei Iancu
>>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>>
>>> On 07.01.2016 11:37, Nabeel wrote:
>>>
>>> Hi Bogdan,
>>>
>>> I used the tshark command as explained here on page 14:
>>> http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_
>>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>>> Mangani-OpenSIPS
>>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>>> _
>>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>>> Summit2015-SIPCapture.pdf
>>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>>>
>>> tshark -o "ssl.desegment_ssl_records: TRUE"  -o
>>> "ssl.desegment_ssl_application_data: TRUE"  -o "ssl.keys_list:
>>> 162.249.6.110,5061,sip,/install/tls/domain.com-key.pem"  -i eth0  -f "tcp
>>> port 5061"
>>>
>>> I'm using a command line version of Linux without a graphic UI, so I
>>> could not "configure Wireshark to decide TLS" as mentioned in that
>>> document, however I did pass the private key in the command as shown above.
>>>
>>> Does tshark require configuring to decode TLS, other than passing the
>>> private key in the command?
>>> Hi Nabeel,
>>>
>>> Indeed, the 408 seems generated by OpenSIPS (after 5 seconds). Such
>>> reply is generated only if the the request was actually sent out (if no
>>> request sent, there is no timeout). But the network capture does not show
>>> anything :( ... maybe wrong capturing ?
>>>
>>> So you see anything in the logs ? have you tried to run with debug level
>>> 4 ?
>>>
>>> Regards,
>>>
>>> Bogdan-Andrei Iancu
>>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>>
>>> On 06.01.2016 23:07, Nabeel wrote:
>>>
>>> I managed to capture the SIP traffic with Wireshark.  It seems that the
>>> party generating the 408 reply is OpenSIPS, not the callee.  OpenSIPS does
>>> not seem to forward the call to the callee at all.
>>>
>>> Below are traces showing a successful call and a call with Request
>>> Timeout.
>>> The server IP is 162.249.6.110, the caller IP is 92.40.249.9, and the
>>> callee IP is 188.29.165.24.
>>>
>>> Trace for a successful call:
>>>
>>> http://pastebin.com/2xn0bkEU
>>>
>>> Trace for a call with Request Timeout:
>>>
>>> http://pastebin.com/WR7BA6pj
>>>
>>> Please advise what may be causing this.
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Request Timeout on INVITE After Idle Registration

[Nabeel]

Hi Bogdan,

I changed log_stderror=yes and log_facility=LOG_DAEMON.  Now I see some
more in the log.  Do you see anything obviously wrong?

http://pastebin.com/MzJW1P1S

On 12 January 2016 at 09:10, Bogdan-Andrei Iancu 
wrote:

> Hi Nabeel,
>
> Be sure you are looking into the right log file - maybe the debug level is
> redirected by your syslog to another log file... Debug level 4 is the most
> verbose one in opensips.
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 08.01.2016 21:04, Nabeel wrote:
>
> Hi Bogdan,
>
> I have the following near the top of my config file:
>
> ## Global Parameters #
>
> debug=4
> log_stderror=no
> log_facility=LOG_LOCAL1
>
> The log I posted earlier is from opensips running with these
> configurations.
> On 8 Jan 2016 3:49 pm, "Bogdan-Andrei Iancu"  wrote:
>
>> Hi Nabeel,
>>
>> have you tried running opensips is debug mode (level 4) to see what it is
>> doing with the request ?
>>
>> Regards,
>>
>> Bogdan-Andrei Iancu
>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>
>> On 07.01.2016 11:37, Nabeel wrote:
>>
>> Hi Bogdan,
>>
>> I used the tshark command as explained here on page 14:
>> http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_
>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>> Mangani-OpenSIPS
>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>> _
>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>> Summit2015-SIPCapture.pdf
>> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>>
>> tshark -o "ssl.desegment_ssl_records: TRUE"  -o
>> "ssl.desegment_ssl_application_data: TRUE"  -o "ssl.keys_list:
>> 162.249.6.110,5061,sip,/install/tls/domain.com-key.pem"  -i eth0  -f "tcp
>> port 5061"
>>
>> I'm using a command line version of Linux without a graphic UI, so I
>> could not "configure Wireshark to decide TLS" as mentioned in that
>> document, however I did pass the private key in the command as shown above.
>>
>> Does tshark require configuring to decode TLS, other than passing the
>> private key in the command?
>> Hi Nabeel,
>>
>> Indeed, the 408 seems generated by OpenSIPS (after 5 seconds). Such reply
>> is generated only if the the request was actually sent out (if no request
>> sent, there is no timeout). But the network capture does not show anything
>> :( ... maybe wrong capturing ?
>>
>> So you see anything in the logs ? have you tried to run with debug level
>> 4 ?
>>
>> Regards,
>>
>> Bogdan-Andrei Iancu
>> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>>
>> On 06.01.2016 23:07, Nabeel wrote:
>>
>> I managed to capture the SIP traffic with Wireshark.  It seems that the
>> party generating the 408 reply is OpenSIPS, not the callee.  OpenSIPS does
>> not seem to forward the call to the callee at all.
>>
>> Below are traces showing a successful call and a call with Request
>> Timeout.
>> The server IP is 162.249.6.110, the caller IP is 92.40.249.9, and the
>> callee IP is 188.29.165.24.
>>
>> Trace for a successful call:
>>
>> http://pastebin.com/2xn0bkEU
>>
>> Trace for a call with Request Timeout:
>>
>> http://pastebin.com/WR7BA6pj
>>
>> Please advise what may be causing this.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Request Timeout on INVITE After Idle Registration

[Nabeel]

Hi Bogdan,

I have the following near the top of my config file:

## Global Parameters #

debug=4
log_stderror=no
log_facility=LOG_LOCAL1

The log I posted earlier is from opensips running with these
configurations.
On 8 Jan 2016 3:49 pm, "Bogdan-Andrei Iancu"  wrote:

> Hi Nabeel,
>
> have you tried running opensips is debug mode (level 4) to see what it is
> doing with the request ?
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 07.01.2016 11:37, Nabeel wrote:
>
> Hi Bogdan,
>
> I used the tshark command as explained here on page 14:
> http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_
> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
> Mangani-OpenSIPS
> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
> _
> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
> Summit2015-SIPCapture.pdf
> <http://opensips.org/pub/events/2015-05-12_OpenSIPS-Summit_Amsterdam/Lorenzo_Mangani-OpenSIPS_Summit2015-SIPCapture.pdf>
>
> tshark -o "ssl.desegment_ssl_records: TRUE"  -o
> "ssl.desegment_ssl_application_data: TRUE"  -o "ssl.keys_list:
> 162.249.6.110,5061,sip,/install/tls/domain.com-key.pem"  -i eth0  -f "tcp
> port 5061"
>
> I'm using a command line version of Linux without a graphic UI, so I could
> not "configure Wireshark to decide TLS" as mentioned in that document,
> however I did pass the private key in the command as shown above.
>
> Does tshark require configuring to decode TLS, other than passing the
> private key in the command?
> Hi Nabeel,
>
> Indeed, the 408 seems generated by OpenSIPS (after 5 seconds). Such reply
> is generated only if the the request was actually sent out (if no request
> sent, there is no timeout). But the network capture does not show anything
> :( ... maybe wrong capturing ?
>
> So you see anything in the logs ? have you tried to run with debug level 4
> ?
>
> Regards,
>
> Bogdan-Andrei Iancu
> OpenSIPS Founder and Developerhttp://www.opensips-solutions.com
>
> On 06.01.2016 23:07, Nabeel wrote:
>
> I managed to capture the SIP traffic with Wireshark.  It seems that the
> party generating the 408 reply is OpenSIPS, not the callee.  OpenSIPS does
> not seem to forward the call to the callee at all.
>
> Below are traces showing a successful call and a call with Request
> Timeout.
> The server IP is 162.249.6.110, the caller IP is 92.40.249.9, and the
> callee IP is 188.29.165.24.
>
> Trace for a successful call:
>
> http://pastebin.com/2xn0bkEU
>
> Trace for a call with Request Timeout:
>
> http://pastebin.com/WR7BA6pj
>
> Please advise what may be causing this.
>
>
>
>
>
>
>
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Request Timeout on INVITE After Idle Registration

[Nabeel]

I managed to capture the SIP traffic with Wireshark.  It seems that the
party generating the 408 reply is OpenSIPS, not the callee.  OpenSIPS does
not seem to forward the call to the callee at all.

Below are traces showing a successful call and a call with Request Timeout.

The server IP is 162.249.6.110, the caller IP is 92.40.249.9, and the
callee IP is 188.29.165.24.

Trace for a successful call:

http://pastebin.com/2xn0bkEU

Trace for a call with Request Timeout:

http://pastebin.com/WR7BA6pj

Please advise what may be causing this.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Request Timeout on INVITE After Idle Registration

[Nabeel]

Hi Bogdan,

I'm trying to use the siptrace module to capture SIP traffic, as I am using
TLS.

In my config file I have the following:

...

loadmodule "siptrace.so"
modparam("siptrace", "db_url", "mysql://user:passwd@host/dbname")
modparam("siptrace", "trace_flag", "TRACE_FLAG")

...

route{
sip_trace();
setflag(TRACE_FLAG);

...

I also did 'opensipsctl fifo sip_trace on', but nothing is captured in the
sip_trace table of the database.

I want to capture all traffic.  What am I missing?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] Request Timeout on INVITE After Idle Registration

[Nabeel]

Hello,

I have an SIP client which works fine during the first 15-20 minutes of
registration.  Registration has an expiry of 24 hours, but after
approximately 15-20 minutes of idle registration, attempts to call the
other phone fails with '408 Request Timeout' error.  Re-registering the
caller's phone makes no improvement to this, but re-registering the
callee's phone allows the calls to go through for another 15-20 minutes.

I understand that Request Timeout can be initiated by either the client or
the server.  The know that the possible reasons are:

1) The callee/endpoint to be reached is somehow set on passive mode and not
responding to any call

2) The server does not get proper information about the location info of
the endpoint after query.

And in the server logs, I only see this:

ACC: call missed:
timestamp=1445625568;method=INVITE;from_tag=z9hG4bK58506340;to_tag=;call_id=879087082370@2a04:4a41:111:cf09::1d92:e0ac%4;code=408;reason=Request
Timeout

However, I am unsure how to troubleshoot this further and how to go about
determining the exact cause.  Any advice or tips would be very helpful.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] 477 Request Failure

[Nabeel]

Hello,

After about 15 minutes of idle registration, call attempts fail with '477
Request Failure'.  In the first 15 minutes after registration, the calls
connect fine.   Please advise on possible cause of this and how to
troubleshoot.  I am using TLS.  Error log is below:

ERROR:core:tcp_connect_blocking: timeout 99438 ms elapsed from 10 s
ERROR:proto_tls:tls_sync_connect: tcp_blocking_connect failed
ERROR:proto_tls:proto_tls_send: connect failed
ERROR:tm:msg_send: send() for proto 3 failed
ERROR:tm:t_forward_nonack: sending request failed
ACC: call missed:
timestamp=1450367086;method=INVITE;from_tag=z9hG4bK14111660;to_tag=;call_id=
491384150620@10.23.233.76;code=477;reason=Request Failure
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] Re-registration / renewal when registration expires

[Nabeel]

Hi,

The registrar module has a 'min_expires' parameter which can force
expiration time on the OpenSIPS side and can override an expiration time
set by the client.  Is there a way for OpenSIPS to force re-registration /
renewal when the registration expires?  Currently, re-registration or
renewal does not occur unless set by the client.  I would like to override
this with an OpenSIPS setting if possible.
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Realtime monitoring of registered end-points

[Nabeel]

Hi,

Are these features enabled by default in the latest version of OpenSIPS
(2.2.x) + NAT script?  Or do they need to be enabled manually?
On 1 Oct 2015 09:33, "Ionut Ionita"  wrote:

> *Hello all,
>
> I’m glad to announce a new feature that allows OpenSIPS to monitor
> (via SIP OPTIONS probing) and disable/delete in realtime the registrations
> which are not responding.
> The previous approach had two issues when came to so called “zombie”
> registrations (registrations which are not valid anymore):
> * resources - such zombie registrations may waste resources in your
> OpenSIPS server (memory,  processing time, DB space, useless NAT pinging,
> TCP connect attempts);
> * user experience - using the zombie registration to reach un-existing
> users translates into useless calls (calls that will simply timeout),
> giving delays in the call setup (instead of going straight to VM, you may
> burn 5-10 seconds in trying to reach the user);
> The main idea  behind these features is to delete contacts that do not
> respond to a certain number of SIP pings. The SIP pinging is provided by
> the
> nathelper module which was enhanced to keep the state of each pinging
> requests
> (basically, the module is waiting and checking the reply of each SIP
> request
> sent to the registered users). For registrations detected as “dead” (not
> responding),
> the nathelper module interacts directly with the usrloc module in order to
> remove the zombie contact.
> For usage perspective, the nathelper module now has two new parameters:
> * ping_threshold - timeout to consider a ping as unanswered;
> * max_pings_lost - the number of unresponded pings after which the
> contact is
> removed from usrloc;
> In order to activate this feature, every contact must have the
> sipping_bflag
> (to be pinged) and remove_on_timeout_bflag (to be deleted on no-answer)
> activated.
> This means these flags must also be configured in the nathelper module. [0]
> For full documentation of the nathelper module including the newly
> added
> feature see [1]. Any feedback is highly appreciated.
>
> Regards,
> Ionut Ionita
>
> [0]http://www.opensips.org/html/docs/modules/2.2.x/nathelper.html#id248011
> [1]http://www.opensips.org/html/docs/modules/2.2.x/nathelper.html
> *
>
> ___
> 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] How to view bitrate, sampling rate and frame size during a SIP call?

[Nabeel]

Hi Jeff,

Thanks for the information.  I checked the SDPs, however mine does not have
the 'a:ptime' line which could indicate the frame size.  Is there a way to
enable this?  Here is an example of what I am seeing:

v=0
> o=user 0 0 IN IP4 162.212.130.252
> s=Session SIP/SDP
> c=IN IP4 162.212.130.252
> t=0 0
> a=ice-ufrag:171m3
> a=ice-pwd:27g6nm2sol7btqvgper41odgjk
> m=audio 55718 RTP/AVP 201 101
> a=rtpmap:201 OPUS/48000
> a=rtpmap:101 telephone-event/8000
> a=fmtp:101 0-15
> a=candidate:1 1 udp 2130706431 10.53.232.161 21000 typ host
> a=candidate:3 1 udp 1694498815 188.29.165.133 49190 typ srflx raddr
> 10.53.232.161 rport 21000
> a=candidate:2 1 udp 16777215 162.212.130.252 55718 typ relay raddr
> 188.29.165.133 rport 49190
> a=candidate:1 2 udp 2130706430 10.53.232.161 21001 typ host
> a=candidate:3 2 udp 1694498814 188.29.165.133 49191 typ srflx raddr
> 10.53.232.161 rport 21001
> a=candidate:2 2 udp 16777214 162.212.130.252 57171 typ relay raddr
> 188.29.165.133 rport 49191



On 8 December 2015 at 14:21, Jeff Pyle  wrote:

> OpenSIPS doesn't handle media so it has no knowledge of these things.  You
> could glean some of this information by inspecting the offer and answer
> SDPs as they pass through.  For example, here is an answer SDP that passed
> through reply_route attached to a 200 OK:
>
> v=0.
> o=FreeSWITCH 1449573019 1449573020 IN IP4 192.168.5.5.
> s=FreeSWITCH.
> c=IN IP4 192.168.5.5.
> t=0 0.
> m=audio 11158 RTP/AVP 9 101.
> a=rtpmap:9 G722/8000.
> a=rtpmap:101 telephone-event/8000.
> a=fmtp:101 0-16.
> a=ptime:20.
>
> From this data you know it's 8khz sampling rate, since it's G722 you know
> it's a 64kbps bitrate, and the ptime is 20ms.  You'd have to account for
> future in-dialog requests (reINVITEs and UPDATEs) that may change these
> parameters.
>
> In order to make this data available for live calls, you'd probably have
> to store them in dialog variables.
>
> In other words, it may be possible to maintain this data from within
> OpenSIPS, but it becomes complicated quickly depending on the variety of
> endpoints and applications you use.  It is generally easier to gather this
> data from the endpoints themselves but you've already said your app does
> not have a way to do that.  That's unfortunate.
>
>
> - Jeff
>
>
> On Sat, Dec 5, 2015 at 11:21 PM, Nabeel  wrote:
>
>> Hello,
>>
>> I need to view the active sampling rate, bitrate and frame size during a
>> SIP call.  The app currently does not have a user interface or custom
>> function to display this.  Is there any other way I can view these
>> parameters during a live call?  What is the simplest way to do this?
>>
>> ___
>> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] How to view bitrate, sampling rate and frame size during a SIP call?

[Nabeel]

Hello,

I need to view the active sampling rate, bitrate and frame size during a
SIP call.  The app currently does not have a user interface or custom
function to display this.  Is there any other way I can view these
parameters during a live call?  What is the simplest way to do this?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Presence server: Using domain name instead of IP address

[Nabeel]

I don't know if that will work, I haven't tried it myself.
On 3 Nov 2015 20:20, "surya"  wrote:

> Oh Thanks this works.
>
> Few comments back you approved similar lines. Nevermind.
> This also means that now I don't need entry in /etc/hosts.
>
> One more question, can I get IP from env into this file?
>
> Thanks,
>
>
>
>
> --
> View this message in context:
> http://opensips-open-sip-server.1449251.n2.nabble.com/Presence-server-Using-domain-name-instead-of-IP-address-tp7599676p7599742.html
> Sent from the OpenSIPS - Users mailing list archive at Nabble.com.
>
> ___
> 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] Presence server: Using domain name instead of IP address

[Nabeel]

You shouldn't need the first line at all.  The 'listen' line should use the
actual IP address:

alias="domain.com"
listen=udp:192.168.192.129:5060
On 3 Nov 2015 20:06, "surya"  wrote:

> Hi Nabeel,
>
> I think I am missing something.
>
> What I did is:
>
> *domain.com="192.168.192.129"*
> alias="domain.com"
> listen=udp:"alias":5060
>
> This gives me syntax error at the first line.
>
> Please correct what I am doing wrong.
>
> Thanks
>
>
>
> --
> View this message in context:
> http://opensips-open-sip-server.1449251.n2.nabble.com/Presence-server-Using-domain-name-instead-of-IP-address-tp7599676p7599740.html
> Sent from the OpenSIPS - Users mailing list archive at Nabble.com.
>
> ___
> 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] Presence server: Using domain name instead of IP address

[Nabeel]

Yes, that looks correct.
On 2 Nov 2015 15:08, "surya"  wrote:

> Will it be look something like below:
>
> alias="domain.com"
>
> listen=udp:alias:5060
> listen=tcp:alias:5060
>
>
>
>
>
> --
> View this message in context:
> http://opensips-open-sip-server.1449251.n2.nabble.com/Presence-server-Using-domain-name-instead-of-IP-address-tp7599676p7599692.html
> Sent from the OpenSIPS - Users mailing list archive at Nabble.com.
>
> ___
> 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] Presence server: Using domain name instead of IP address

[Nabeel]

There is no field.   It can be added near the listening IP address line.
On 2 Nov 2015 14:45, "surya"  wrote:

> Hi Nabeel,
>
> Is there a field somewhere as *alias*? I am in version 1.8.3 .
>
>
>
> --
> View this message in context:
> http://opensips-open-sip-server.1449251.n2.nabble.com/Presence-server-Using-domain-name-instead-of-IP-address-tp7599676p7599690.html
> Sent from the OpenSIPS - Users mailing list archive at Nabble.com.
>
> ___
> 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] Presence server: Using domain name instead of IP address

[Nabeel]

I also use alias="domain.com" in my config file.   Without it my SIP client
did not connect to the server with domain in SIP URI.
On 2 Nov 2015 08:10, "surya"  wrote:

> Hi All,
>
>
> I am trying to use a domain name instead of IP address but not sure of the
> proper way.
>
> 1st I changed the SIP_DOMAIN in the opensipctlrc file, but it doesn't seem
> to have any effect.
>
> Then, in opensips.cfg I replaced all IPs with domain.com and added an
> entry
> in etc hosts. This seems to be working, but somehow I feel this is not the
> proper way.
>
> *sample:
> /listen=udp:opensipstest.org:5060   # CUSTOMIZE ME/*
>
> I'll appreciate if someone can tell the proper way to use a domain name in
> the presence server.
>
>
>
> --
> View this message in context:
> http://opensips-open-sip-server.1449251.n2.nabble.com/Presence-server-Using-domain-name-instead-of-IP-address-tp7599676.html
> Sent from the OpenSIPS - Users mailing list archive at Nabble.com.
>
> ___
> 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] TLS Module load Error Opensips 2.1

[Nabeel]

In my experience, version 1.11.5 had TLS enabled by default but version 2.x
had to be compiled with TLS explicitly enabled at compile time.
On 31 Oct 2015 08:47, "John Mathew"  wrote:

> Yes.
> OpenSIPS is trying to load the module, but it is unable to find libssl
> file. But that file is in Linux lib folder and is accessible.
>
> On the same machine OpenSIPS 1.11.5 with TLS is working perfectly. So ssl
> package is correct.
> My doubt is OpenSIPS looking for libssl in a wrong folder.
> On Oct 31, 2015 14:06, "Nabeel"  wrote:
>
>> Did you compile Opensips with TLS enabled or is it enabled by default?
>> On 31 Oct 2015 08:32, "John Mathew"  wrote:
>>
>>> File path in the config is correct. All other modules successfully
>>> loaded. Only TLS is having the issue.
>>> On Oct 31, 2015 13:42, "Nabeel"  wrote:
>>>
>>>> Seems like a file path issue.  Check path to modules folder.
>>>>
>>>> On 31 October 2015 at 07:32, John Mathew 
>>>> wrote:
>>>>
>>>>> I have tried installing in both Ubuntu and Centos 6.4. Same issue on
>>>>> both platform.
>>>>> I think this is some sort of bug in Opensips 2.1.
>>>>>
>>>>> On Sat, Oct 31, 2015 at 12:48 PM, John Mathew <
>>>>> john.mat...@divoxmedia.com> wrote:
>>>>>
>>>>>> In config that is already given.
>>>>>> On Oct 31, 2015 12:11, "Hamid R. Hashmi" 
>>>>>> wrote:
>>>>>>
>>>>>>> Load module proto_tls.
>>>>>>>
>>>>>>> On 31 Oct 2015 10:58, John Mathew 
>>>>>>> wrote:
>>>>>>> >
>>>>>>> > Hi,
>>>>>>> >
>>>>>>> > Anybody knows why this happens when loading tls module in opensips
>>>>>>> 2.1?
>>>>>>> > Openssl ans openssl-devel is install.
>>>>>>> > libssl files are in /usr/lib64
>>>>>>> >
>>>>>>> > Oct 31 01:23:44 SERVER24 opensips: ERROR:core:sr_load_module:
>>>>>>> could not open module : proto_tls.so: cannot open shared
>>>>>>> object file: No such file or directory
>>>>>>> > Oct 31 01:23:44 SERVER24 opensips: ERROR:core:load_module: failed
>>>>>>> to load module
>>>>>>> > Oct 31 01:23:44 SERVER24 opensips: CRITICAL:core:yyerror: parse
>>>>>>> error in config file /usr/local/opensips/etc/opensips/opensips.cfg, line
>>>>>>> 30, column 13-14: failed to load module proto_tls.so#012
>>>>>>> >
>>>>>>> >
>>>>>>> > Thanks.
>>>>>>> ___
>>>>>>> Users mailing list
>>>>>>> Users@lists.opensips.org
>>>>>>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> John Mathew   Divox International Inc. | Divox FZ LLC
>>>>>  +971-7-243-1145
>>>>>  +91-9037-11
>>>>>  john.mat...@divoxmedia.com 
>>>>>  www.divoxmedia.com
>>>>>  375 Park Avenue, Seagram Building, Suite: 2607, New York City, New
>>>>> York, USA - 10152
>>>>> <https://www.linkedin.com/pub/john-mathew/20/634/642>
>>>>>
>>>>> ___
>>>>> 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
>>>>
>>>>
>>> ___
>>> 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
>>
>>
> ___
> 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] TLS Module load Error Opensips 2.1

[Nabeel]

Did you compile Opensips with TLS enabled or is it enabled by default?
On 31 Oct 2015 08:32, "John Mathew"  wrote:

> File path in the config is correct. All other modules successfully loaded.
> Only TLS is having the issue.
> On Oct 31, 2015 13:42, "Nabeel"  wrote:
>
>> Seems like a file path issue.  Check path to modules folder.
>>
>> On 31 October 2015 at 07:32, John Mathew 
>> wrote:
>>
>>> I have tried installing in both Ubuntu and Centos 6.4. Same issue on
>>> both platform.
>>> I think this is some sort of bug in Opensips 2.1.
>>>
>>> On Sat, Oct 31, 2015 at 12:48 PM, John Mathew <
>>> john.mat...@divoxmedia.com> wrote:
>>>
>>>> In config that is already given.
>>>> On Oct 31, 2015 12:11, "Hamid R. Hashmi" 
>>>> wrote:
>>>>
>>>>> Load module proto_tls.
>>>>>
>>>>> On 31 Oct 2015 10:58, John Mathew  wrote:
>>>>> >
>>>>> > Hi,
>>>>> >
>>>>> > Anybody knows why this happens when loading tls module in opensips
>>>>> 2.1?
>>>>> > Openssl ans openssl-devel is install.
>>>>> > libssl files are in /usr/lib64
>>>>> >
>>>>> > Oct 31 01:23:44 SERVER24 opensips: ERROR:core:sr_load_module: could
>>>>> not open module : proto_tls.so: cannot open shared object
>>>>> file: No such file or directory
>>>>> > Oct 31 01:23:44 SERVER24 opensips: ERROR:core:load_module: failed to
>>>>> load module
>>>>> > Oct 31 01:23:44 SERVER24 opensips: CRITICAL:core:yyerror: parse
>>>>> error in config file /usr/local/opensips/etc/opensips/opensips.cfg, line
>>>>> 30, column 13-14: failed to load module proto_tls.so#012
>>>>> >
>>>>> >
>>>>> > Thanks.
>>>>> ___
>>>>> Users mailing list
>>>>> Users@lists.opensips.org
>>>>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>>>>
>>>>
>>>
>>>
>>> --
>>> John Mathew   Divox International Inc. | Divox FZ LLC
>>>  +971-7-243-1145
>>>  +91-9037-11
>>>  john.mat...@divoxmedia.com 
>>>  www.divoxmedia.com
>>>  375 Park Avenue, Seagram Building, Suite: 2607, New York City, New
>>> York, USA - 10152
>>> <https://www.linkedin.com/pub/john-mathew/20/634/642>
>>>
>>> ___
>>> 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
>>
>>
> ___
> 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] TLS Module load Error Opensips 2.1

[Nabeel]

Seems like a file path issue.  Check path to modules folder.

On 31 October 2015 at 07:32, John Mathew  wrote:

> I have tried installing in both Ubuntu and Centos 6.4. Same issue on both
> platform.
> I think this is some sort of bug in Opensips 2.1.
>
> On Sat, Oct 31, 2015 at 12:48 PM, John Mathew 
> wrote:
>
>> In config that is already given.
>> On Oct 31, 2015 12:11, "Hamid R. Hashmi"  wrote:
>>
>>> Load module proto_tls.
>>>
>>> On 31 Oct 2015 10:58, John Mathew  wrote:
>>> >
>>> > Hi,
>>> >
>>> > Anybody knows why this happens when loading tls module in opensips 2.1?
>>> > Openssl ans openssl-devel is install.
>>> > libssl files are in /usr/lib64
>>> >
>>> > Oct 31 01:23:44 SERVER24 opensips: ERROR:core:sr_load_module: could
>>> not open module : proto_tls.so: cannot open shared object
>>> file: No such file or directory
>>> > Oct 31 01:23:44 SERVER24 opensips: ERROR:core:load_module: failed to
>>> load module
>>> > Oct 31 01:23:44 SERVER24 opensips: CRITICAL:core:yyerror: parse error
>>> in config file /usr/local/opensips/etc/opensips/opensips.cfg, line 30,
>>> column 13-14: failed to load module proto_tls.so#012
>>> >
>>> >
>>> > Thanks.
>>> ___
>>> Users mailing list
>>> Users@lists.opensips.org
>>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>>
>>
>
>
> --
> John Mathew   Divox International Inc. | Divox FZ LLC
>  +971-7-243-1145
>  +91-9037-11
>  john.mat...@divoxmedia.com 
>  www.divoxmedia.com
>  375 Park Avenue, Seagram Building, Suite: 2607, New York City, New York,
> USA - 10152
> 
>
> ___
> 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

[OpenSIPS-Users] ERROR:core:tcp_connect_blocking: timeout

[Nabeel]

Hi,

On some calls, I'm getting this timeout error in the OpenSIPS log even when
the call successfully connects and the other phone rings.  How can it be
avoided?

INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 25
> ERROR:core:tcp_connect_blocking: timeout 99231 ms elapsed from 10 s
> ERROR:proto_tls:tls_sync_connect: tcp_blocking_connect failed
> ERROR:proto_tls:proto_tls_send: connect failed
> ERROR:tm:msg_send: send() for proto 3 failed
> ERROR:tm:t_forward_nonack: sending request failed
> incoming reply
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Does Media_Relay component implies in impossible direct media?

[Nabeel]

ICE in general uses STUN (RFC 5245), however I do not use Mediaproxy myself
so cannot comment on how it handles STUN.   Someone with Mediaproxy may be
able to answer.
On 19 Oct 2015 14:41, "Rodrigo Pimenta Carvalho"  wrote:

> Ok Nabeel.
>
>
> In my environment, I intend to have MediaProxy and all clients supporting
> ICE.
>
> However, in my environment/network doesn't exist a STUN server.
>
>
> So, my last questions (before I start the MediaProxy installation
> procedures) about this subject are:
>
>
> 1 -  Does MediaProxy work as a STUN server, or just as a TURN server?
>
>
> 2 - Is it always necessary to provide a STUN server separately, even using
> the MediaProxy, if direct media is desired?
>
>
> 3 - Let's suppose I will not provide a separately STUN server, in this
> case will the media always pass through the media relay,?
>
>
> Thank you again!
>
> Best regards.
>
>
>
> RODRIGO PIMENTA CARVALHO
> Inatel Competence Center
> Software
> Ph: +55 35 3471 9200 RAMAL 979
>
>
> --
> *De:* users-boun...@lists.opensips.org 
> em nome de Nabeel 
> *Enviado:* segunda-feira, 19 de outubro de 2015 10:40
> *Para:* OpenSIPS users mailling list
> *Assunto:* Re: [OpenSIPS-Users] Does Media_Relay component implies in
> impossible direct media?
>
> Yes, that is how it should work.  ICE tries to connect directly first
> before trying other methods.  If direct connection fails, then it tries
> STUN, then media relay.  The clients have to support ICE too.
>
> On 19 October 2015 at 11:35, Rodrigo Pimenta Carvalho 
> wrote:
>
>> Hi Nabeel.
>>
>>
>> Thank you very much for your answer.
>>
>> If I understood it well, while using MediaProxy, the clients may connect
>> directly, but they may not too. ICE will decide it for me and I don't need
>> to worry about it.
>>
>> Tell me if I got it, please.
>>
>>
>> Best regards.
>>
>>
>>
>>
>> RODRIGO PIMENTA CARVALHO
>> Inatel Competence Center
>> Software
>> Ph: +55 35 3471 9200 RAMAL 979
>>
>>
>> --
>> *De:* users-boun...@lists.opensips.org 
>> em nome de Nabeel 
>> *Enviado:* sábado, 17 de outubro de 2015 19:56
>> *Para:* OpenSIPS users mailling list
>> *Assunto:* Re: [OpenSIPS-Users] Does Media_Relay component implies in
>> impossible direct media?
>>
>>
>> Rodrigo,
>>
>> It is the responsibility of ICE to determine in which situation it is
>> best to establish a direct connection, versus use of STUN, versus use of a
>> media relay.
>>
>> We cannot manually decide which clients will connect directly and which
>> will not.
>> On 17 Oct 2015 22:07, "Rodrigo Pimenta Carvalho" 
>> wrote:
>>
>>>
>>> Hi.
>>>
>>>
>>> In my project, UACs should run direct media. RTP packets direct
>>> transmitted from each other.
>>>
>>> However, due to NAT presence, I'm thinking about to use MediaProxy.
>>>
>>>
>>> In this case, is it possible to get direct media after some UAC
>>> negotiations passing through the Media_Relay component, or is it
>>> impossible? That is, does the media will always pass through the
>>> Media_Relay host too?
>>>
>>>
>>> Any hint will be very helpful!
>>>
>>> Thanks alot.
>>>
>>>
>>>
>>>
>>>
>>> RODRIGO PIMENTA CARVALHO
>>> Inatel Competence Center
>>> Software
>>> Ph: +55 35 3471 9200 RAMAL 979
>>>
>>> ___
>>> 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
>>
>>
>
> ___
> 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] Does Media_Relay component implies in impossible direct media?

[Nabeel]

Yes, that is how it should work.  ICE tries to connect directly first
before trying other methods.  If direct connection fails, then it tries
STUN, then media relay.  The clients have to support ICE too.

On 19 October 2015 at 11:35, Rodrigo Pimenta Carvalho 
wrote:

> Hi Nabeel.
>
>
> Thank you very much for your answer.
>
> If I understood it well, while using MediaProxy, the clients may connect
> directly, but they may not too. ICE will decide it for me and I don't need
> to worry about it.
>
> Tell me if I got it, please.
>
>
> Best regards.
>
>
>
>
> RODRIGO PIMENTA CARVALHO
> Inatel Competence Center
> Software
> Ph: +55 35 3471 9200 RAMAL 979
>
>
> ------
> *De:* users-boun...@lists.opensips.org 
> em nome de Nabeel 
> *Enviado:* sábado, 17 de outubro de 2015 19:56
> *Para:* OpenSIPS users mailling list
> *Assunto:* Re: [OpenSIPS-Users] Does Media_Relay component implies in
> impossible direct media?
>
>
> Rodrigo,
>
> It is the responsibility of ICE to determine in which situation it is best
> to establish a direct connection, versus use of STUN, versus use of a media
> relay.
>
> We cannot manually decide which clients will connect directly and which
> will not.
> On 17 Oct 2015 22:07, "Rodrigo Pimenta Carvalho" 
> wrote:
>
>>
>> Hi.
>>
>>
>> In my project, UACs should run direct media. RTP packets direct
>> transmitted from each other.
>>
>> However, due to NAT presence, I'm thinking about to use MediaProxy.
>>
>>
>> In this case, is it possible to get direct media after some UAC
>> negotiations passing through the Media_Relay component, or is it
>> impossible? That is, does the media will always pass through the
>> Media_Relay host too?
>>
>>
>> Any hint will be very helpful!
>>
>> Thanks alot.
>>
>>
>>
>>
>>
>> RODRIGO PIMENTA CARVALHO
>> Inatel Competence Center
>> Software
>> Ph: +55 35 3471 9200 RAMAL 979
>>
>> ___
>> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Does Media_Relay component implies in impossible direct media?

[Nabeel]

Rodrigo,

It is the responsibility of ICE to determine in which situation it is best
to establish a direct connection, versus use of STUN, versus use of a media
relay.

We cannot manually decide which clients will connect directly and which
will not.
On 17 Oct 2015 22:07, "Rodrigo Pimenta Carvalho"  wrote:

>
> Hi.
>
>
> In my project, UACs should run direct media. RTP packets direct
> transmitted from each other.
>
> However, due to NAT presence, I'm thinking about to use MediaProxy.
>
>
> In this case, is it possible to get direct media after some UAC
> negotiations passing through the Media_Relay component, or is it
> impossible? That is, does the media will always pass through the
> Media_Relay host too?
>
>
> Any hint will be very helpful!
>
> Thanks alot.
>
>
>
>
>
> RODRIGO PIMENTA CARVALHO
> Inatel Competence Center
> Software
> Ph: +55 35 3471 9200 RAMAL 979
>
> ___
> 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] What choise? MediaProxy module or RTPProxy module?

[Nabeel]

Rtpproxy is only able to relay media (indirectly) through the server, so
this can't achieve what you want.

Some users have said Mediaproxy has ICE capability.   If that is so, then
ICE should be able to negotiate direct communication between some clients.
On 15 Oct 2015 18:36, "Rodrigo Pimenta Carvalho"  wrote:

>
> Hi.
>
>
> Today I'm searching for a solution that allows me to use OpenSIPS, SIP
> over TCP, end-nodes behind NATs and direct media.
>
>
> As someone pointed, I'm reading now about MediaProxy module.
>
>
> However, if I'm right, by using MediaProxy I will get only media relay.
> That is, MediaProxy does only provide media relay. Am I right?
>
>
> I need an way to implement direct media between the end-nodes, behind
> NATs. I want to avoid pass the media through the MediaProxy, due to
> performance reasons.
>
>
> Should I give up using MediaProxy and move to RTPProxy? Would it be a good
> decision?
>
>
> When it is better to use MediaProxy, and when it is better to use
> RTPProxy?
>
>
> Any hint will be very helpful!
>
>
> Best regards.
>
>
>
> RODRIGO PIMENTA CARVALHO
> Inatel Competence Center
> Software
> Ph: +55 35 3471 9200 RAMAL 979
>
> ___
> 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

[OpenSIPS-Users] Bad port in uri

[Nabeel]

Hi,

I'm getting the followinh error with IPv6 when attempting to make a call.
Is it related to the SIP client or OpenSIPS?

09-28 03:02:33.564  18214-21831/com.sipdomain I/IntegratedSipProvider﹕
message:
SIP/2.0 500 Internal Error
Via: SIP/2.0/TLS
[2a04:4a41:218:c8d2::2383:20a5]:34248;received=94.197.121.62;rport=7085;branch=z9hG4bK99966
To: ;tag=22aedaebbea315632eb770d97d0e8a0d.6772
From: ;tag=z9hG4bK61460932
Call-ID: 651586519504@2a04:4a41:218:c8d2::2383:20a5%4
CSeq: 2 INVITE





ERROR:core:parse_uri: bad port in uri (error at char a in state 8) parsed:
(24)
/
(69)
ERROR:registrar:update_contacts: failed to parse contact

ERROR:core:parse_uri: bad port in uri (error at char a in state 8) parsed:
(24)
/
(69)
ERROR:nathelper:get_contact_uri: failed to parse Contact URI
ERROR:core:parse_uri: bad port in uri (error at char a in state 8) parsed:
(24)
/
(69)
ERROR:nathelper:get_contact_uri: failed to parse Contact URI
ERROR:core:parse_uri: bad port in uri (error at char a in state 8) parsed:
(24)
/
(69)
ERROR:core:parse_sip_msg_uri: bad uri 
ERROR:tm:new_t: uri invalid
ERROR:tm:t_newtran: new_t failed
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

Thanks to a suggestion from Aron, I have now solved this problem.  The file
causing the permission error requires the '/usr/local' prefix in its
filepath, which the service file doesn't have.  The PID filepath is also
incorrect and the EnvironmentFile value should be '-/etc/default/opensips',
not '-/etc/sysconfig/opensips' like in the fedora package.

I am posting my working service file below for CentOS and I hope this gets
included in the future releases of OpenSIPS.

[Unit]
> Description=OpenSIPS is a very fast and flexible SIP (RFC3261) server
> After=network.target mariadb.service


> [Service]
> Type=forking
> User=root
> Group=root
> EnvironmentFile=-/etc/default/opensips
> PIDFile=/var/run/opensips.pid
> ExecStart=/usr/local/sbin/opensips -P /var/run/opensips.pid -f
> /usr/local/etc/opensips/opensips.cfg $OPTIONS
> ExecStartPre=/usr/local/sbin/opensips -c -f
> /usr/local/etc/opensips/opensips.cfg
> Restart=always
> TimeoutStopSec=30s
> [Install]
> WantedBy=multi-user.target
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

No, I just compiled and installed OpenSIPS in the normal manner while
logged in as root.

On 20 September 2015 at 06:00, Podrigal, Aron 
wrote:

> Are you running this on a container / chrooted env?
>
> On Sun, Sep 20, 2015 at 12:39 AM, Nabeel  wrote:
>
>> This is what it says:
>>
>>
>>> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: Starting OpenSIPS is
>>> a very fast and flexible SIP (RFC3261) server...
>>> -- Subject: Unit opensips.service has begun with start-up
>>> -- Defined-By: systemd
>>> -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
>>> --
>>> -- Unit opensips.service has begun starting up.
>>> Sep 20 05:21:45 server1.sipdomain.com systemd[4022]: Failed at step
>>> EXEC spawning /usr/sbin/opensips: Permission denied
>>> -- Subject: Process /usr/sbin/opensips could not be executed
>>> -- Defined-By: systemd
>>> -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
>>> --
>>> -- The process /usr/sbin/opensips could not be executed and failed.
>>> --
>>> -- The error number returned while executing this process is 13.
>>> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: opensips.service:
>>> control process exited, code=exited status=203
>>> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: Failed to start
>>> OpenSIPS is a very fast and flexible SIP (RFC3261) server.
>>> -- Subject: Unit opensips.service has failed
>>> -- Defined-By: systemd
>>> -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
>>> --
>>> -- Unit opensips.service has failed.
>>> --
>>> -- The result is failed.
>>> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: Unit opensips.service
>>> entered failed state.
>>
>>
>>
>> On 20 September 2015 at 05:12, Podrigal, Aron 
>> wrote:
>>
>>> Do you see anything informative in your journal after trying to start
>>> opensips?
>>>
>>> sudo journalctl -xn 350
>>>
>>> On Sat, Sep 19, 2015 at 11:46 PM, Nabeel 
>>> wrote:
>>>
>>>> Please see the strace output at these links:
>>>>
>>>> http://pastebin.com/G1Uv5s1E
>>>>
>>>> http://pastebin.com/gnesw4tW
>>>>
>>>> On 20 September 2015 at 04:30, Podrigal, Aron >>> > wrote:
>>>>
>>>>> Can you post the strace output by running
>>>>>
>>>>> sudo strace -ff -o /tmp/opensips_strace_output systemctl start
>>>>> opensips.service
>>>>> cat /tmp/opensips_strace_output*
>>>>>
>>>>> That will help see where the error occurs.
>>>>>
>>>>> On Sat, Sep 19, 2015 at 11:21 PM, Nabeel 
>>>>> wrote:
>>>>>
>>>>>> Why can't the service file be automatically generated at the time of
>>>>>> installation/compilation on the relevant OS?
>>>>>> It seems to be quite a basic feature missing.
>>>>>>
>>>>>> On 20 September 2015 at 04:16, Nabeel 
>>>>>> wrote:
>>>>>>
>>>>>>> Unfortunately, changing the user/group to 'root' did not work either.
>>>>>>>
>>>>>>> The same error occurs and OpenSIPS fails to start:
>>>>>>>
>>>>>>> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>>>>>>>
>>>>>>>
>>>>>>>  Hasn't anyone else seen this on CentOS?
>>>>>>>
>>>>>>>
>>>>>>> On 18 September 2015 at 22:21, sevpal  wrote:
>>>>>>>
>>>>>>>> If you change User= and Group= in the service file to root, you
>>>>>>>> might not see such a message. To run as user opensips, then user 
>>>>>>>> opensips
>>>>>>>> must exist on the system; in addition, ownership on files/folders that 
>>>>>>>> will
>>>>>>>> be accessed by Opensips needs to be changed.
>>>>>>>> *From:* Nabeel 
>>>>>>>> *Sent:* Tuesday, September 15, 2015 9:47 PM
>>>>>>>> *To:* OpenSIPS users mailling list 
>>>>>>>> *Subject:* Re: [OpenSIPS-Users] Restart=always and
>>>>>>>> After=mysql.service
>>>>>>>>
>>>>&

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

This is what it says:


> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: Starting OpenSIPS is a
> very fast and flexible SIP (RFC3261) server...
> -- Subject: Unit opensips.service has begun with start-up
> -- Defined-By: systemd
> -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
> --
> -- Unit opensips.service has begun starting up.
> Sep 20 05:21:45 server1.sipdomain.com systemd[4022]: Failed at step EXEC
> spawning /usr/sbin/opensips: Permission denied
> -- Subject: Process /usr/sbin/opensips could not be executed
> -- Defined-By: systemd
> -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
> --
> -- The process /usr/sbin/opensips could not be executed and failed.
> --
> -- The error number returned while executing this process is 13.
> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: opensips.service:
> control process exited, code=exited status=203
> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: Failed to start
> OpenSIPS is a very fast and flexible SIP (RFC3261) server.
> -- Subject: Unit opensips.service has failed
> -- Defined-By: systemd
> -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
> --
> -- Unit opensips.service has failed.
> --
> -- The result is failed.
> Sep 20 05:21:45 server1.sipdomain.com systemd[1]: Unit opensips.service
> entered failed state.



On 20 September 2015 at 05:12, Podrigal, Aron 
wrote:

> Do you see anything informative in your journal after trying to start
> opensips?
>
> sudo journalctl -xn 350
>
> On Sat, Sep 19, 2015 at 11:46 PM, Nabeel  wrote:
>
>> Please see the strace output at these links:
>>
>> http://pastebin.com/G1Uv5s1E
>>
>> http://pastebin.com/gnesw4tW
>>
>> On 20 September 2015 at 04:30, Podrigal, Aron 
>> wrote:
>>
>>> Can you post the strace output by running
>>>
>>> sudo strace -ff -o /tmp/opensips_strace_output systemctl start
>>> opensips.service
>>> cat /tmp/opensips_strace_output*
>>>
>>> That will help see where the error occurs.
>>>
>>> On Sat, Sep 19, 2015 at 11:21 PM, Nabeel 
>>> wrote:
>>>
>>>> Why can't the service file be automatically generated at the time of
>>>> installation/compilation on the relevant OS?
>>>> It seems to be quite a basic feature missing.
>>>>
>>>> On 20 September 2015 at 04:16, Nabeel  wrote:
>>>>
>>>>> Unfortunately, changing the user/group to 'root' did not work either.
>>>>>
>>>>> The same error occurs and OpenSIPS fails to start:
>>>>>
>>>>> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>>>>>
>>>>>
>>>>>  Hasn't anyone else seen this on CentOS?
>>>>>
>>>>>
>>>>> On 18 September 2015 at 22:21, sevpal  wrote:
>>>>>
>>>>>> If you change User= and Group= in the service file to root, you might
>>>>>> not see such a message. To run as user opensips, then user opensips must
>>>>>> exist on the system; in addition, ownership on files/folders that will be
>>>>>> accessed by Opensips needs to be changed.
>>>>>> *From:* Nabeel 
>>>>>> *Sent:* Tuesday, September 15, 2015 9:47 PM
>>>>>> *To:* OpenSIPS users mailling list 
>>>>>> *Subject:* Re: [OpenSIPS-Users] Restart=always and
>>>>>> After=mysql.service
>>>>>>
>>>>>> I tried that service file on CentOS 7 but got the following error
>>>>>> when trying to start the service:
>>>>>>
>>>>>> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>>>>>>
>>>>>>
>>>>>> Changing permissions of that directory did not solve the problem:
>>>>>>
>>>>>> chmod +x /usr/sbin/opensips
>>>>>>
>>>>>> chmod -R 777 /usr/sbin/opensips
>>>>>>
>>>>>> Does the service file need modification for CentOS?
>>>>>>
>>>>>>
>>>>>> On 7 September 2015 at 19:14, Nabeel  wrote:
>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> Most of my services have a .service file located at
>>>>>>> /usr/lib/systemd/system where I can set 'Restart=always' or
>>>>>>> 'After=mysql.service' / 'After=mariadb.

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

Please see the strace output at these links:

http://pastebin.com/G1Uv5s1E

http://pastebin.com/gnesw4tW

On 20 September 2015 at 04:30, Podrigal, Aron 
wrote:

> Can you post the strace output by running
>
> sudo strace -ff -o /tmp/opensips_strace_output systemctl start
> opensips.service
> cat /tmp/opensips_strace_output*
>
> That will help see where the error occurs.
>
> On Sat, Sep 19, 2015 at 11:21 PM, Nabeel  wrote:
>
>> Why can't the service file be automatically generated at the time of
>> installation/compilation on the relevant OS?
>> It seems to be quite a basic feature missing.
>>
>> On 20 September 2015 at 04:16, Nabeel  wrote:
>>
>>> Unfortunately, changing the user/group to 'root' did not work either.
>>>
>>> The same error occurs and OpenSIPS fails to start:
>>>
>>> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>>>
>>>
>>>  Hasn't anyone else seen this on CentOS?
>>>
>>>
>>> On 18 September 2015 at 22:21, sevpal  wrote:
>>>
>>>> If you change User= and Group= in the service file to root, you might
>>>> not see such a message. To run as user opensips, then user opensips must
>>>> exist on the system; in addition, ownership on files/folders that will be
>>>> accessed by Opensips needs to be changed.
>>>> *From:* Nabeel 
>>>> *Sent:* Tuesday, September 15, 2015 9:47 PM
>>>> *To:* OpenSIPS users mailling list 
>>>> *Subject:* Re: [OpenSIPS-Users] Restart=always and After=mysql.service
>>>>
>>>> I tried that service file on CentOS 7 but got the following error when
>>>> trying to start the service:
>>>>
>>>> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>>>>
>>>>
>>>> Changing permissions of that directory did not solve the problem:
>>>>
>>>> chmod +x /usr/sbin/opensips
>>>>
>>>> chmod -R 777 /usr/sbin/opensips
>>>>
>>>> Does the service file need modification for CentOS?
>>>>
>>>>
>>>> On 7 September 2015 at 19:14, Nabeel  wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> Most of my services have a .service file located at
>>>>> /usr/lib/systemd/system where I can set 'Restart=always' or
>>>>> 'After=mysql.service' / 'After=mariadb.service' to make sure the server
>>>>> starts at the appropriate times.
>>>>>
>>>>> However, there is no such .service file for OpenSIPS.  Please advise
>>>>> how to create this .service file for OpenSIPS.  I heard about 'respawn' 
>>>>> but
>>>>> don't know how exactly to use this and would prefer to use a .service file
>>>>> like other services.
>>>>>
>>>>
>>>>
>>>> --
>>>> ___
>>>> 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
>>>>
>>>>
>>>
>>
>> ___
>> Users mailing list
>> Users@lists.opensips.org
>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>
>>
>
>
> --
> Aron Podrigal
> -
> '101', '1110010', '110', '1101110'   '101', '110',
> '1100100', '1110010', '1101001', '1100111', '111', '1101100'
>
> P: '2b', '31', '33', '34', '37', '34', '35', '38', '36', '30', '39', '39'
>
>
> ___
> 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] Restart=always and After=mysql.service

[Nabeel]

Why can't the service file be automatically generated at the time of
installation/compilation on the relevant OS?
It seems to be quite a basic feature missing.

On 20 September 2015 at 04:16, Nabeel  wrote:

> Unfortunately, changing the user/group to 'root' did not work either.
>
> The same error occurs and OpenSIPS fails to start:
>
> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>
>
>  Hasn't anyone else seen this on CentOS?
>
>
> On 18 September 2015 at 22:21, sevpal  wrote:
>
>> If you change User= and Group= in the service file to root, you might not
>> see such a message. To run as user opensips, then user opensips must exist
>> on the system; in addition, ownership on files/folders that will be
>> accessed by Opensips needs to be changed.
>> *From:* Nabeel 
>> *Sent:* Tuesday, September 15, 2015 9:47 PM
>> *To:* OpenSIPS users mailling list 
>> *Subject:* Re: [OpenSIPS-Users] Restart=always and After=mysql.service
>>
>> I tried that service file on CentOS 7 but got the following error when
>> trying to start the service:
>>
>> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>>
>>
>> Changing permissions of that directory did not solve the problem:
>>
>> chmod +x /usr/sbin/opensips
>>
>> chmod -R 777 /usr/sbin/opensips
>>
>> Does the service file need modification for CentOS?
>>
>>
>> On 7 September 2015 at 19:14, Nabeel  wrote:
>>
>>> Hi,
>>>
>>> Most of my services have a .service file located at
>>> /usr/lib/systemd/system where I can set 'Restart=always' or
>>> 'After=mysql.service' / 'After=mariadb.service' to make sure the server
>>> starts at the appropriate times.
>>>
>>> However, there is no such .service file for OpenSIPS.  Please advise how
>>> to create this .service file for OpenSIPS.  I heard about 'respawn' but
>>> don't know how exactly to use this and would prefer to use a .service file
>>> like other services.
>>>
>>
>>
>> --
>> ___
>> 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
>>
>>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

Unfortunately, changing the user/group to 'root' did not work either.

The same error occurs and OpenSIPS fails to start:

Failed at step EXEC spawning /usr/sbin/opensips: Permission denied


 Hasn't anyone else seen this on CentOS?


On 18 September 2015 at 22:21, sevpal  wrote:

> If you change User= and Group= in the service file to root, you might not
> see such a message. To run as user opensips, then user opensips must exist
> on the system; in addition, ownership on files/folders that will be
> accessed by Opensips needs to be changed.
> *From:* Nabeel 
> *Sent:* Tuesday, September 15, 2015 9:47 PM
> *To:* OpenSIPS users mailling list 
> *Subject:* Re: [OpenSIPS-Users] Restart=always and After=mysql.service
>
> I tried that service file on CentOS 7 but got the following error when
> trying to start the service:
>
> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>
>
> Changing permissions of that directory did not solve the problem:
>
> chmod +x /usr/sbin/opensips
>
> chmod -R 777 /usr/sbin/opensips
>
> Does the service file need modification for CentOS?
>
>
> On 7 September 2015 at 19:14, Nabeel  wrote:
>
>> Hi,
>>
>> Most of my services have a .service file located at
>> /usr/lib/systemd/system where I can set 'Restart=always' or
>> 'After=mysql.service' / 'After=mariadb.service' to make sure the server
>> starts at the appropriate times.
>>
>> However, there is no such .service file for OpenSIPS.  Please advise how
>> to create this .service file for OpenSIPS.  I heard about 'respawn' but
>> don't know how exactly to use this and would prefer to use a .service file
>> like other services.
>>
>
>
> --
> ___
> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

Hi,

Does anyone have a working service file for CentOS 7?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

I have selinux disabled, but to test that suggestion I temporarily enabled
selinux and restored context, but the error was not resolved.
On 16 Sep 2015 03:14, "Podrigal, Aron"  wrote:

> Does restoring your selinux context solve the problem.
>
> chcon --type=bin_t /usr/sbin/opensips
>
> If that helps make it permanent by
> restorecon -R -v /usr/sbin/opensips
> On Sep 15, 2015 9:47 PM, "Nabeel"  wrote:
>
>> I tried that service file on CentOS 7 but got the following error when
>> trying to start the service:
>>
>> Failed at step EXEC spawning /usr/sbin/opensips: Permission denied
>>
>>
>> Changing permissions of that directory did not solve the problem:
>>
>> chmod +x /usr/sbin/opensips
>>
>> chmod -R 777 /usr/sbin/opensips
>>
>> Does the service file need modification for CentOS?
>>
>>
>> On 7 September 2015 at 19:14, Nabeel  wrote:
>>
>>> Hi,
>>>
>>> Most of my services have a .service file located at
>>> /usr/lib/systemd/system where I can set 'Restart=always' or
>>> 'After=mysql.service' / 'After=mariadb.service' to make sure the server
>>> starts at the appropriate times.
>>>
>>> However, there is no such .service file for OpenSIPS.  Please advise how
>>> to create this .service file for OpenSIPS.  I heard about 'respawn' but
>>> don't know how exactly to use this and would prefer to use a .service file
>>> like other services.
>>>
>>
>>
>> ___
>> 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
>
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] Restart=always and After=mysql.service

[Nabeel]

I tried that service file on CentOS 7 but got the following error when
trying to start the service:

Failed at step EXEC spawning /usr/sbin/opensips: Permission denied


Changing permissions of that directory did not solve the problem:

chmod +x /usr/sbin/opensips

chmod -R 777 /usr/sbin/opensips

Does the service file need modification for CentOS?


On 7 September 2015 at 19:14, Nabeel  wrote:

> Hi,
>
> Most of my services have a .service file located at
> /usr/lib/systemd/system where I can set 'Restart=always' or
> 'After=mysql.service' / 'After=mariadb.service' to make sure the server
> starts at the appropriate times.
>
> However, there is no such .service file for OpenSIPS.  Please advise how
> to create this .service file for OpenSIPS.  I heard about 'respawn' but
> don't know how exactly to use this and would prefer to use a .service file
> like other services.
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Re: [OpenSIPS-Users] ha1 password authentication

[Nabeel]

Hi razvan,

By default, the opensips script from menuconfig already has 'calculate_ha1'
set to 'yes' (and 'password_column' set to 'password').  Do you mean set
'calculate_ha1' to '0'?  Then should the 'password_column' remain
'password', or should it point to the 'ha1' column?  If the column remains
'password', should this column contain the encoded ha1 password?

Please clarify.
On 14 Sep 2015 08:04, "Răzvan Crainea"  wrote:

> Hi, Nabeel!
>
> By default, OpenSIPS uses the password column to authenticate users. If
> you want to use HA1 authentication, you should set the calculate_ha1
> parameter[1].
>
> [1] http://www.opensips.org/html/docs/modules/2.1.x/auth_db#id293468
>
> Best regards,
>
> Răzvan Crainea
> OpenSIPS Solutionswww.opensips-solutions.com
>
> On 09/13/2015 10:25 AM, Nabeel wrote:
>
> Just tested this and I'm not able to register without having the plain
> text password present in the 'password' column in db.  How can I make
> OpenSIPS register using only the ha1 hashed value? I tried commenting out
> and changing the 'password_column' attribute in config but it didn't work.
> On 12 Sep 2015 12:40, "Nabeel"  wrote:
>
>> Hi,
>>
>> My SIP client only uses the 'ha1' password field from database to
>> authenticate users and ignores the 'password' field totally.  Will OpenSIPS
>> allow such authentication to complete and register the user by using only
>> the ha1 hashed password?
>>
>
>
> ___
> Users mailing 
> listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>
>
> ___
> 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] ha1 password authentication

[Nabeel]

Just tested this and I'm not able to register without having the plain text
password present in the 'password' column in db.  How can I make OpenSIPS
register using only the ha1 hashed value? I tried commenting out and
changing the 'password_column' attribute in config but it didn't work.
On 12 Sep 2015 12:40, "Nabeel"  wrote:

> Hi,
>
> My SIP client only uses the 'ha1' password field from database to
> authenticate users and ignores the 'password' field totally.  Will OpenSIPS
> allow such authentication to complete and register the user by using only
> the ha1 hashed password?
>
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

[OpenSIPS-Users] ha1 password authentication

[Nabeel]

Hi,

My SIP client only uses the 'ha1' password field from database to
authenticate users and ignores the 'password' field totally.  Will OpenSIPS
allow such authentication to complete and register the user by using only
the ha1 hashed password?
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users