Re: [asterisk-users] PJSIP signaling question

2016-02-29 Thread George Joseph 
On Mon, Feb 29, 2016 at 2:04 PM, Kevin Long 
wrote:

>
>
> Greetings.
>
>
> I am using the PJSIP driver with TLS transport, and my endpoints are SIP
> mobile apps operating in environments that I do not control.
>
>  I would like Asterisk to default to sending INVITES and all other SIP
> signals to endpoints via the existing SIP TLS connection which is already
> established, rather than trying to create a new TLS connection to an
> endpoint which is likely behind a NAT which will not allow a new inbound
> TCP/TLS connection.
>
>
> My experience with chan_sip suggest to me that this was the default
> behavior, or more likely a fallback behavior, because I never had this
> issue before with endpoints not receiving INVITES so long as they were
> registered and had an open SIP control connection.
>
>
> I thought that I could avoid these failed outbound connections by
> commenting out the “transport” option on my endpoint configurations, but
> tcpdump is showing me that asterisk is still trying to create *new* TLS
> outbound connections to my endpoints, which are failing.
>
>
>
>
This was actually an issue in pjproject which I just fixed last week. :)

It's in pjproject "trunk" so you'll have to download and build it from
their subversion repository.

Now whether you use "transport=" or not, pjproject will look for an
existing connection to the remote endpoint before attempting to create a
new one.

I tested it with the current Asterisk 13 branch and I *think* it'll work
with recent Asterisk releases as well.  If it doesn't, let me know.



> Thank you for your time
>
> Kevin
>
>
> -
>
>
>
>
> My simple pjsip config file:
>
>
>
>
>
> [transport-tls]
> type=transport
> protocol=tls
> bind=0.0.0.0:5061
> local_net=10.50.55.0/24
> external_media_address=x.x.x.x
> external_signaling_address=x.x.x.x
> cert_file=/etc/asterisk/keys/dev1.crt
> priv_key_file=/etc/asterisk/keys/dev1.key
> ca_list_file=/etc/asterisk/keys/ca.crt
> cipher=AES256-SHA
> method=tlsv1
>
> ;===EXTENSION 6001
>
> [6000]
> type=endpoint
> context=internal
> disallow=all
> allow=ulaw
> ;transport=transport-tls
> auth=auth6000
> aors=6000
> direct_media=no
> rewrite_contact=yes  ; necessary if endpoint does not know/register public
> ip:port
> ice_support=no
> force_rport=yes
> rtp_symmetric=yes
> media_encryption=sdes
>
>
> [auth6000]
> type=auth
> auth_type=userpass
> password=6000
> username=6000
>
> [6000]
> type=aor
> max_contacts=1
> remove_existing=yes
>
>
> ;===EXTENSION 6001
>
> [6001]
> type=endpoint
> context=internal
> disallow=all
> allow=ulaw
> ;transport=transport-tls
> auth=auth6001
> aors=6001
> direct_media=no
> rewrite_contact=yes  ; necessary if endpoint does not know/register public
> ip:port
> ice_support=no
> force_rport=yes
> rtp_symmetric=yes
> media_encryption=sdes
>
>
>
> [auth6001]
> type=auth
> auth_type=userpass
> password=6001
> username=6001
>
> [6001]
> type=aor
> max_contacts=1
> remove_existing=yes
> --
> _
> -- Bandwidth and Colocation Provided by http://www.api-digital.com --
> New to Asterisk? Join us for a live introductory webinar every Thurs:
>http://www.asterisk.org/hello
>
> asterisk-users mailing list
> To UNSUBSCRIBE or update options visit:
>http://lists.digium.com/mailman/listinfo/asterisk-users
>
-- 
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
   http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

[asterisk-users] PJSIP signaling question

2016-02-29 Thread Kevin Long 


Greetings.


I am using the PJSIP driver with TLS transport, and my endpoints are SIP mobile 
apps operating in environments that I do not control. 

 I would like Asterisk to default to sending INVITES and all other SIP signals 
to endpoints via the existing SIP TLS connection which is already established, 
rather than trying to create a new TLS connection to an endpoint which is 
likely behind a NAT which will not allow a new inbound TCP/TLS connection.


My experience with chan_sip suggest to me that this was the default behavior, 
or more likely a fallback behavior, because I never had this issue before with 
endpoints not receiving INVITES so long as they were registered and had an open 
SIP control connection.


I thought that I could avoid these failed outbound connections by commenting 
out the “transport” option on my endpoint configurations, but tcpdump is 
showing me that asterisk is still trying to create *new* TLS outbound 
connections to my endpoints, which are failing.




Thank you for your time

Kevin


-




My simple pjsip config file:





[transport-tls]
type=transport
protocol=tls
bind=0.0.0.0:5061
local_net=10.50.55.0/24
external_media_address=x.x.x.x
external_signaling_address=x.x.x.x
cert_file=/etc/asterisk/keys/dev1.crt
priv_key_file=/etc/asterisk/keys/dev1.key
ca_list_file=/etc/asterisk/keys/ca.crt
cipher=AES256-SHA
method=tlsv1
 
;===EXTENSION 6001
 
[6000]
type=endpoint
context=internal
disallow=all
allow=ulaw
;transport=transport-tls
auth=auth6000
aors=6000
direct_media=no
rewrite_contact=yes  ; necessary if endpoint does not know/register public 
ip:port
ice_support=no
force_rport=yes
rtp_symmetric=yes
media_encryption=sdes


[auth6000]
type=auth
auth_type=userpass
password=6000
username=6000
 
[6000]
type=aor
max_contacts=1
remove_existing=yes


;===EXTENSION 6001

[6001]
type=endpoint
context=internal
disallow=all   
allow=ulaw
;transport=transport-tls
auth=auth6001
aors=6001
direct_media=no
rewrite_contact=yes  ; necessary if endpoint does not know/register public 
ip:port
ice_support=no
force_rport=yes
rtp_symmetric=yes
media_encryption=sdes



[auth6001]
type=auth
auth_type=userpass
password=6001
username=6001

[6001]
type=aor
max_contacts=1
remove_existing=yes

smime.p7s
Description: S/MIME cryptographic signature
-- 
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
   http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] Asterisk 13 and WebRTC. Is wiki page still valid ?

2016-02-29 Thread Olivier 
2016-02-19 12:01 GMT+01:00 Marek Červenka :

> on my own server
>

Today, I'm back from holidays trip.

First of all, thanks for replying !

I'll try to use jssip as you suggested.

Anyway, I'm still failing to understand if wiki's page [1] is still valid
with Asterisk 13, and if it's not valid anymore, which is the main change
that prevent things to work.

[1] https://wiki.asterisk.org/wiki/display/AST/WebRTC+tutorial+using+SIPML5


>
> i want try jssip
> https://github.com/versatica/JsSIP
> it looks like a lot  "livelier" than sipml5
>
> any experience with jssip?
>
>
> Dne 18.2.2016 v 16:01 Olivier napsal(a):
>
>
>
> 2016-02-18 15:42 GMT+01:00 Marek Červenka :
>
>> my experience with pjsip for webrtc
>>
>> http://lists.digium.com/pipermail/asterisk-users/2015-September/287562.html
>>
>>
>> Yes I saw this post earlier today.
> Having to fight 14 days scared me a bit !
>
> Did you set sipml5 on your own server or did you use Live demo (
> https://www.doubango.org/sipml5/call.htm?svn=241) ?
>
>
>
>> Dne 18.2.2016 v 15:36 Olivier napsal(a):
>>
>>
>>
>> 2016-02-18 14:57 GMT+01:00 Simon Hohberg <
>> simon.hohb...@mcs-datalabs.com>:
>>
>>>
>>> Is it implied here that both HTTPS and WSS must also come from the same
 server (Same Origin Policy) ?

>>> No, the same origin policy does not apply to web sockets.
>>>
>>> Then, can I also install my own WebRTC demo page on my own private
 Asterisk server and access this demo page through HTTPS ?
 If I'm not mistaken, this should fulfill all requirements.

>>> It doesn't matter where the asterisk server is hosted. It is important
>>> where the web application comes from. If you don't want to use https and
>>> wss you only have the option to host the web app locally (on the same
>>> machine as the browser that loads the page), which probably makes sense
>>> only for development. Otherwise you have to use https and wss for the
>>> reasons discussed earlier.
>>>
>>> Hope it helps.
>>
>>
>>
>> At least, it helped me to realize I still have several more things to
>> learn ;-)
>>
>> My setup is the following:
>> - an asterisk server,
>> - a PC,
>> - asterisk server and PC are installed on the same LAN
>> - sipM5 live demo outside my LAN
>> - no NAT/PAT configuration allowing incoming communications from the
>> outside.
>>
>> Is using sipML live demo as a way to rapidly test private Asterisk WebRTC
>> capabilies, something achievable ?
>> What would keep this from working ?
>>
>>
> --
> ---
> Marek Cervenka
> ===
>
>
> --
> _
> -- Bandwidth and Colocation Provided by http://www.api-digital.com --
> New to Asterisk? Join us for a live introductory webinar every Thurs:
>http://www.asterisk.org/hello
>
> asterisk-users mailing list
> To UNSUBSCRIBE or update options visit:
>http://lists.digium.com/mailman/listinfo/asterisk-users
>
-- 
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
   http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

[asterisk-users] Zoiper on Windows Phone

2016-02-29 Thread Vitor Mazuco 
Hello everyone, I have some problems to enable push the Zoiper
Windows Phone in my Asterisk 11.


Below is the result of CLI

 == Using SIP RTP CoS mark 5
-- Executing [1033@ramais:1] Answer("SIP/1030-0201", "") in new stack
   > 0x7efc90024190 -- Probation passed - setting RTP source
address to 179.XX.XXX.XX:57741
[Feb 29 12:32:28] NOTICE[4348][C-01ce]: res_rtp_asterisk.c:4441
ast_rtp_read: Unknown RTP codec 95 received from '179.XX.XX.XX:57741'
   > 0x7efc90024190 -- Probation passed - setting RTP source
address to 179.XX.XX.XX:57741
-- Executing [1033@ramais:2] Set("SIP/1030-0201", "location=")
in new stack
-- Executing [1033@ramais:3] Verbose("SIP/1030-0201", "0,
getting push info  ") in new stack
 getting push info
-- Executing [1033@ramais:4] Set("SIP/1030-0201",
"regx="X-PUSH-URI=([0-9a-zA-Z\.\:\/\_]+)"") in new stack
-- Executing [1033@ramais:5] Set("SIP/1030-0201", "push=") in new stack
-- Executing [1033@ramais:6] System("SIP/1030-0201",
"/usr/bin/push.sh ") in new stack
-- Executing [1033@ramais:7] Wait("SIP/1030-0201", "1") in new stack
-- Executing [1033@ramais:8] Dial("SIP/1030-0201", "SIP/1033")
in new stack
[Feb 29 12:32:29] WARNING[4348][C-01ce]: app_dial.c:2437
dial_exec_full: Unable to create channel of type 'SIP' (cause 20 -
Subscriber absent)
  == Everyone is busy/congested at this time (1:0/0/1)
-- Auto fallthrough, channel 'SIP/1030-0201' status is 'CHANUNAVAIL'
asterisk*CLI>

I've created the file more push.sh qualification in the dialplan. But the
Windows Phone can not run on Asterisk.

Does anyone know another method for this?

Thanks in advanced.

-- 
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
   http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] Crash asterisk res_odbc

2016-02-29 Thread Maxime 

Hi,

Thank you for the reply.

My OS is : Debian 7.

But i have more than 20 servers with the same features/resources (OS, 
material, ... ) without the issue.


Regards.

Le 28/02/2016 21:49, Leandro Dardini a écrit :


Which operating system are you using? I have experienced the same 
problem on several OS except for CentOS 6. I suppose an ODBC problem 
on newer OS version.


Leandro

Il 24/Feb/2016 05:30 PM, "Maxime" > ha scritto:


Dear list,

i have a issue

Asterisk crash (Module res_odbc exactly) after the same log who is
"/ERROR[23805] astobj2.c: bad magic number.../"
you will see on the log :

Today

[2016-02-24 16:00:38] ERROR[23805] *astobj2.c: bad magic number
0x552f302e for 0x7fe3505b3958*
[2016-02-24 16:00:44] Asterisk 11.2-cert1 built by root @
Voice_server on a x86_64 running Linux on 2013-04-09 14:16:57 UTC
[2016-02-24 16:00:44] NOTICE[31321] loader.c: 2 modules will be
loaded.
[2016-02-24 16:00:44] NOTICE[31321] res_odbc.c: Connecting asterisk
[2016-02-24 16:00:44] NOTICE[31321] res_odbc.c: res_odbc:
Connected to asterisk [MySQL-asterisk]
[2016-02-24 16:00:44] NOTICE[31321] res_odbc.c: Registered ODBC
class 'asterisk' dsn->[MySQL-asterisk]

Yesterday :

[2016-02-23 15:59:12] ERROR[19824] *astobj2.c: bad magic number
0x20 for 0x27a5558*
[2016-02-23 15:59:18] Asterisk 11.2-cert1 built by root @
Voice_server on a x86_64 running Linux on 2013-04-09 14:16:57 UTC
[2016-02-23 15:59:18] NOTICE[23791] loader.c: 2 modules will be
loaded.
[2016-02-23 15:59:18] NOTICE[23791] res_odbc.c: Connecting asterisk
[2016-02-23 15:59:18] NOTICE[23791] res_odbc.c: res_odbc:
Connected to asterisk [MySQL-asterisk]
[2016-02-23 15:59:18] NOTICE[23791] res_odbc.c: Registered ODBC
class 'asterisk' dsn->[MySQL-asterisk]

Effect : many trunk sip are down during few minutes
Oddness : same hours

On google i found many times  "memory corruption was the
assumption" ...

Have you ever seen this kind of problem ?

thank you in advance

Version : Asterisk 11.2-cert1
Os : Debian 7-64

-- 


Maxime

--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users





-- 
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
   http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users