opensipstack-devel

Re: [OpenSIPStack] ATLSIP and ntdll.dll crashing

[Yacine A. <yac...@ms...>]

Hello,I'm getting the same problem even with Windows XP Pro SP2 and for me =
it works well on two laptops running on both Home edition and Media center =
Edition.Is it possible that ATLSIP can have incompatibility with processors=
?Because i tested it on 6 Celeron PCs and two P4 running Windows XP PRO and=
 it always crashing after some calls.> Date: Fri, 10 Aug 2007 15:19:37 +080=
0> From: ip...@so...> To: ope...@li...urceforg=
e.net> Subject: Re: [OpenSIPStack] ATLSIP and  ntdll.dll crashing> > Hi Whi=
t,> > I'm on a XP Pro box today so I can't reproduce this yet.  How are you=
 > deploying your softphone to XP Home? If it's via a VS Setup project, try=
 > packaging it with the appropriate CRT and ATL merge modules (and the > p=
olicies that come with them).> > Regards,> Ilian> > Whit Thiele wrote:> > H=
ey Guys,> >> > I was going to post this on the ATLSIP list, but it bounced =
back to me.> > (Maybe not working yet?)> >> >> >> > Anyway, I've been using=
 ATLSIP for some time now, with pretty good success.> > I'm currently havin=
g a very bizarre issue. I've built a C# softphone and it> > runs fine on XP=
 Pro machines. However, when it's installed on XP Home boxes,> > it sporadi=
cally crashes, giving the windows Send/Don't Send big report.> >> > The mod=
ule which is affected and causing the crash is "ntdll.dll" . The> > softpho=
ne typically works for 2-3 calls, then crashes. Has anyone> > experienced t=
his? I've compiled ATLSIP from CVS yesterday. Any suggestions> > on how to =
debug or even a possible solution for this would be great!> >> >> > Whit> >=
> >> >> >> >> > -----------------------------------------------------------=
--------------> > This SF.net email is sponsored by: Splunk Inc.> > Still g=
repping through log files to find problems?  Stop.> > Now Search log events=
 and configuration files using AJAX and a browser.> > Download your FREE co=
py of Splunk now >>  http://get.splunk.com/> > ____________________________=
___________________> > opensipstack-devel mailing list> > opensipstack-deve=
l...@li...> > https://lists.sourceforge.net/lists/listinfo/ope=
nsipstack-devel> >> >> >   > > > ------------------------------------------=
-------------------------------> This SF.net email is sponsored by: Splunk =
Inc.> Still grepping through log files to find problems?  Stop.> Now Search=
 log events and configuration files using AJAX and a browser.> Download you=
r FREE copy of Splunk now >>  http://get.splunk.com/> _____________________=
__________________________> opensipstack-devel mailing list> opensipstack-d=
ev...@li...> https://lists.sourceforge.net/lists/listinfo/op=
ensipstack-devel
_________________________________________________________________
David Guetta a r=E9uni les sons les plus connus de Messenger dans le Mix Me=
ssenger, le son de l=92=E9t=E9 ! T=E9l=E9chargez-le gratuitement !
http://specials.divertissements.fr.msn.com/mixmessenger=

Re: [OpenSIPStack] ATLSIP and ntdll.dll crashing

[Ilian J. C. P. <ip...@so...>]

Hi Whit,

I'm on a XP Pro box today so I can't reproduce this yet.  How are you 
deploying your softphone to XP Home? If it's via a VS Setup project, try 
packaging it with the appropriate CRT and ATL merge modules (and the 
policies that come with them).

Regards,
Ilian

Whit Thiele wrote:
> Hey Guys,
>
> I was going to post this on the ATLSIP list, but it bounced back to me.
> (Maybe not working yet?)
>
>
>
> Anyway, I've been using ATLSIP for some time now, with pretty good success.
> I'm currently having a very bizarre issue. I've built a C# softphone and it
> runs fine on XP Pro machines. However, when it's installed on XP Home boxes,
> it sporadically crashes, giving the windows Send/Don't Send big report.
>
> The module which is affected and causing the crash is "ntdll.dll" . The
> softphone typically works for 2-3 calls, then crashes. Has anyone
> experienced this? I've compiled ATLSIP from CVS yesterday. Any suggestions
> on how to debug or even a possible solution for this would be great!
>
>
> Whit
>
>
>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> opensipstack-devel mailing list
> ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensipstack-devel
>
>
>   

Re: [OpenSIPStack] is it necessary to parse RTP in openSBC

[Ashish K. <ash...@gm...>]

Hi Baclor,

Thanks for the reply.
Is there any design document about openSBC, which will tell me in detail
about how it is implementing the NAtting/ ALG functinality and how it will
handle the Media streams.
For Call  Transfer, we will use the relay approach.
Also, if i want to just relay the Media packets, can you let me know the
algorithm you have applied in the openSBC.

Also, in openSBC product, is High Availability supported or it is in roadmap
?


On 8/10/07, Joegen E. Baclor <joe...@gm...> wrote:
>
> inline...
>
>
> Ashish Khare wrote:
> > Hi Baclor,
> > This is still not clear to me.
> > Lets take a example:
> > Sip Client A is talking to Sip Client B through Proxy/B2BUA(P) which
> > handles only SIP signaling messages.
> > Now in Call, Sip Client A is trasnferred to Sip CLient C and now B and
> > C are talking.
> > But still they are abel to talk.
> > Then how this case is different from yours. Can  you please elaborate
> > and explain to me.
>
> There are two ways OpenSBC handles REFER.  The default is to relay the
> REFER to the UA and let the UA do the transfer request.   This is ok
> because the UA knows that there will be a change in the audio session.
> The second way (Local REFER) will not relay the REFER.   Instead OpenSBC
> do the transfer.   This leaves the other UA to not know that the call is
> actually transfered.   If the transfer succeeded, a new media would with
> a different SSRC would have been created.   If OpenSBC just relays that,
> the UA may reject the packets because the ssrc has already changed.
>
>
> >
> > we are considering to build ALG. We have our own SIP stack ( Proxy and
> > B2BUA ) but we dont have RTP stack.  Also we dont want to parse the
> > RTP stream. Just Rely it, based on source and destination IP and
> > ports. Is this feasible ?. We are also exploring your openSBC if we
> > can used it.
>
>
>
> Of course this is feasible.  You will have to change some lines of code
> in the media interface but it wont take much.    Just post questions
> about the code if you need to clarify something.
>
> >
> >
> >
> >
>
>

Re: [OpenSIPStack] errorInfo given by Event_OutgoingCallRejected

[GCC <vir...@in...>]

Thanks for your reply. It does work!!

By the way, some parts of the code use signals-and-slots mechanism of Qt library, so that's where the keyword emit come from.

Re: [OpenSIPStack] ATLSIP and ntdll.dll crashing

[Andre S. <eds...@ya...>]

Build your softphone on XP Home and make an installer for it.

Whit Thiele <wh...@wh...> wrote:  

Hey Guys,

I was going to post this on the ATLSIP list, but it bounced back to me.
(Maybe not working yet?)



Anyway, I've been using ATLSIP for some time now, with pretty good success.
I'm currently having a very bizarre issue. I've built a C# softphone and it
runs fine on XP Pro machines. However, when it's installed on XP Home boxes,
it sporadically crashes, giving the windows Send/Don't Send big report.

The module which is affected and causing the crash is "ntdll.dll" . The
softphone typically works for 2-3 calls, then crashes. Has anyone
experienced this? I've compiled ATLSIP from CVS yesterday. Any suggestions
on how to debug or even a possible solution for this would be great!


Whit





-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
opensipstack-devel mailing list
ope...@li...
https://lists.sourceforge.net/lists/listinfo/opensipstack-devel


       
---------------------------------
Building a website is a piece of cake. 
Yahoo! Small Business gives you all the tools to get online.

[OpenSIPStack] ATLSIP and ntdll.dll crashing

[Whit T. <wh...@wh...>]

Hey Guys,

I was going to post this on the ATLSIP list, but it bounced back to me.
(Maybe not working yet?)



Anyway, I've been using ATLSIP for some time now, with pretty good success.
I'm currently having a very bizarre issue. I've built a C# softphone and it
runs fine on XP Pro machines. However, when it's installed on XP Home boxes,
it sporadically crashes, giving the windows Send/Don't Send big report.

The module which is affected and causing the crash is "ntdll.dll" . The
softphone typically works for 2-3 calls, then crashes. Has anyone
experienced this? I've compiled ATLSIP from CVS yesterday. Any suggestions
on how to debug or even a possible solution for this would be great!


Whit

Re: [OpenSIPStack] is it necessary to parse RTP in openSBC

[Joegen E. B. <joe...@gm...>]

inline...

Ashish Khare wrote:
> Hi Baclor,
> This is still not clear to me.
> Lets take a example:
> Sip Client A is talking to Sip Client B through Proxy/B2BUA(P) which 
> handles only SIP signaling messages.
> Now in Call, Sip Client A is trasnferred to Sip CLient C and now B and 
> C are talking.
> But still they are abel to talk.
> Then how this case is different from yours. Can  you please elaborate 
> and explain to me.

There are two ways OpenSBC handles REFER.  The default is to relay the 
REFER to the UA and let the UA do the transfer request.   This is ok 
because the UA knows that there will be a change in the audio session.   
The second way (Local REFER) will not relay the REFER.   Instead OpenSBC 
do the transfer.   This leaves the other UA to not know that the call is 
actually transfered.   If the transfer succeeded, a new media would with 
a different SSRC would have been created.   If OpenSBC just relays that, 
the UA may reject the packets because the ssrc has already changed.


>  
> we are considering to build ALG. We have our own SIP stack ( Proxy and 
> B2BUA ) but we dont have RTP stack.  Also we dont want to parse the 
> RTP stream. Just Rely it, based on source and destination IP and 
> ports. Is this feasible ?. We are also exploring your openSBC if we 
> can used it.



Of course this is feasible.  You will have to change some lines of code 
in the media interface but it wont take much.    Just post questions 
about the code if you need to clarify something.

>  
>  
>
>  

Re: [OpenSIPStack] is it necessary to parse RTP in openSBC

[Ashish K. <ash...@gm...>]

Hi Baclor,
This is still not clear to me.
Lets take a example:
Sip Client A is talking to Sip Client B through Proxy/B2BUA(P) which handles
only SIP signaling messages.
Now in Call, Sip Client A is trasnferred to Sip CLient C and now B and C are
talking.
But still they are abel to talk.
Then how this case is different from yours. Can  you please elaborate and
explain to me.

we are considering to build ALG. We have our own SIP stack ( Proxy and B2BUA
) but we dont have RTP stack.  Also we dont want to parse the RTP stream.
Just Rely it, based on source and destination IP and ports. Is this feasible
?. We are also exploring your openSBC if we can used it.




On 8/9/07, Joegen E. Baclor <joe...@gm...> wrote:
>
> There are cases where the SSRC may change in mid call.  One example is
> when called is transferred and OpenSBC is configured for Local REFER.
> There are UAs that will ignore packets when SSRC is changed.  Creating
> new packets gets rid of this problem.
>
>
> Ashish Khare wrote:
> > Hi,
> >
> > I am a new user of openSBC product.
> > I was just going through it and have a quiestion.
> >
> > Why it is  necessary to parse RTP. Why cannot we use only the IP and
> port in
> > the source and destination fiels to route the messages. why such type of
> > details are needed to be known by
> > ssrc=490290042 ntp=2007/8/8-2:22:0.995224 rtp=76800 psent=2 osent=320
> > ssrc=490290042 fraction=0 lost=0 last_seq=131074 jitter=0
> > lsr=419:09:26.072dlsr=0
> >
> -------------------------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc.
> > Still grepping through log files to find problems?  Stop.
> > Now Search log events and configuration files using AJAX and a browser.
> > Download your FREE copy of Splunk now >>  http://get.splunk.com/
> > _______________________________________________
> > opensipstack-devel mailing list
> > ope...@li...
> > https://lists.sourceforge.net/lists/listinfo/opensipstack-devel
> >
> >
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> opensipstack-devel mailing list
> ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensipstack-devel
>

Re: [OpenSIPStack] errorInfo given by Event_OutgoingCallRejected

[Joegen E. B. <joe...@gm...>]

This is a race condition.  I am not familiar with the "emit" keyword but 
by the looks of it, this keyword triggers an event using another thread 
or an event loop.  Since you gave the event just a pointer to the string 
errorInfo, your event MAY get garbage if  errorInfo is already out of 
scope or has been deleted in the underlying code of opensipstack before 
your event actually gets a time-slice.   I suggest you memcpy() the 
c_str() to a char * before passing it to your event.  Of course, you 
need to make sure that your code free()s the string when you are done 
with it.

GCC wrote:
> Hello,
>
>   When I try to display the information of the object errorInfo, I sometimes get some weird characters. The following is how I did it:
>
> ==================================================
>
> void  LoginDialog::Event_OutgoingCallRejected( 
>   int errorCode, 
>   const OString & errorInfo 
> )
> { 
>
>  emit outCallRejected( errorInfo.c_str() );
>
>  } 
>
> And I use the expression:
>
> connect( this, SIGNAL( outCallRejected( const char * ) ), this, SLOT( setCallFail( const char * ) ) );
>
> ===============================================
>
> void LoginDialog::setCallFail( const char *eInfo )
> {
>  OStringStream tempEI; 
>
>  tempEI << "> Error: " << eInfo << "\n";
>  m_OverallStatus->textCursor().insertText( tempEI.str().c_str() );
>
> }
>
>  
>
> The same problem seems to occur in other pure virtual functions. The screenshot of the problem is here:
>
> http://img249.imageshack.us/img249/8875/weirdpn3.jpg
>
> Can anyone help me fix it?
>
> Regards,
>
> GCC
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> opensipstack-devel mailing list
> ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensipstack-devel
>
>
>   

[OpenSIPStack] errorInfo given by Event_OutgoingCallRejected

[GCC <vir...@in...>]

Hello,

  When I try to display the information of the object errorInfo, I sometimes get some weird characters. The following is how I did it:

==================================================

void  LoginDialog::Event_OutgoingCallRejected( 
  int errorCode, 
  const OString & errorInfo 
)
{ 

 emit outCallRejected( errorInfo.c_str() );

 } 

And I use the expression:

connect( this, SIGNAL( outCallRejected( const char * ) ), this, SLOT( setCallFail( const char * ) ) );

===============================================

void LoginDialog::setCallFail( const char *eInfo )
{
 OStringStream tempEI; 

 tempEI << "> Error: " << eInfo << "\n";
 m_OverallStatus->textCursor().insertText( tempEI.str().c_str() );

}

 

The same problem seems to occur in other pure virtual functions. The screenshot of the problem is here:

http://img249.imageshack.us/img249/8875/weirdpn3.jpg

Can anyone help me fix it?

Regards,

GCC

Re: [OpenSIPStack] Record phone call

[tomach <to...@dg...>]

Ok, great I am looking forward for any news about it.


Good luck!

[OpenSIPStack] [OpenSBC] : Open SBC in openIMScore network

[Kolneath S. <kol...@sa...>]

Hi all,

Any of you could tell me wat to do to Open SBC in order to be compatible=20
in IMS network.
Does it need to modify the code or not ie. more header to add,...etc

Thanks in advance for your help.

Best regards,
Kolneath SOMETH
kol...@sa...
kol...@en...
Mobile :  +33 (0)6 18 17 32 45



" Ce courriel et les documents qui y sont attaches peuvent contenir des inf=
ormations confidentielles. Si vous n'etes  pas le destinataire escompte, me=
rci d'en informer l'expediteur immediatement et de detruire ce courriel  ai=
nsi que tous les documents attaches de votre systeme informatique. Toute di=
vulgation, distribution ou copie du present courriel et des documents attac=
hes sans autorisation prealable de son emetteur est interdite."=20

" This e-mail and any attached documents may contain confidential or propri=
etary information. If you are not the intended recipient, please advise the=
 sender immediately and delete this e-mail and all attached documents from =
your computer system. Any unauthorised disclosure, distribution or copying =
hereof is prohibited."

Re: [OpenSIPStack] Record phone call

[Joegen E. B. <joe...@gm...>]

There will be a bit of a delay for this feature.  I've been tide up with 
other more mundane tasks in the office.  However, I've already uploaded 
the mixer class OpalMixer.cxx in CVS.  The only thing left is using this 
class to grab and mix PCM samples from the media channels.   I'll make 
an announcement as soon as it is ready.

tomach wrote:
> Hello Joegen,
>
> How are things going?
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> opensipstack-devel mailing list
> ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensipstack-devel
>
>   

Re: [OpenSIPStack] is it necessary to parse RTP in openSBC

[Joegen E. B. <joe...@gm...>]

There are cases where the SSRC may change in mid call.  One example is 
when called is transferred and OpenSBC is configured for Local REFER.  
There are UAs that will ignore packets when SSRC is changed.  Creating 
new packets gets rid of this problem.


Ashish Khare wrote:
> Hi,
>
> I am a new user of openSBC product.
> I was just going through it and have a quiestion.
>
> Why it is  necessary to parse RTP. Why cannot we use only the IP and port in
> the source and destination fiels to route the messages. why such type of
> details are needed to be known by
> ssrc=490290042 ntp=2007/8/8-2:22:0.995224 rtp=76800 psent=2 osent=320
> ssrc=490290042 fraction=0 lost=0 last_seq=131074 jitter=0
> lsr=419:09:26.072dlsr=0
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> opensipstack-devel mailing list
> ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensipstack-devel
>
>   

Re: [OpenSIPStack] Dtmf Detection (RFC 2833) in OpenSBC

[Joegen E. B. <joe...@gm...>]

It would help if you let us know what exactly it is you are trying to 
achieve and why you want to detect DTMF in openSBC RTP stream. You are 
correct OpenSBC does not process DTMF because it does not have any use 
for it as a B2BUA. However, you can try calling the media server. The 
default number for the Media Server is 5000@opensbc_address. The media 
sever detects your DTMF. It can be your basis how you can detect DTMF in 
OpenSBC B2BUA RTP Stream.


Gaurav Kheterpal wrote:
>
> Hello,
>
> I’m facing an issue with RTP on OpenSBC 1.1.4 in capturing RFC 2833 
> packets. My network configuration is:-
>
> Xlite UA (Outbound) (.64) ß----à OpenSBC (.115) ß-------à Xlite UA 
> (Inbound) (.60)
>
> * Outbound UA is registered on OpenSBC
>
> * Inbound UA is registered on OpenSBC
>
> * A route exists from OpenSBC to Xlite UA (Inbound) (.60)
>
> RTP_Session::SendReceiveStatus RTP_Session::OnReceiveData(const 
> RTP_DataFrame & frame) is the function where all rtp packets get 
> processed.
>
> The source file rfc2833.cxx has the function 
> OpalRFC2833Proto::ReceivedPacket() which seems to be there for 
> receiving an RFC2833 packet but I could not find a reference from 
> where it’s being invoked. Is this a bug in OpenSBC?
>
> Attached are the level 5 logs for inbound and outbound call where both 
> UAs tried to send & receive DTMF as RFC 2833 packets.
>
> I’ll be thankful for any guidance on what’s going wrong.
>
> Regards,
> Gaurav
>

[OpenSIPStack] is it necessary to parse RTP in openSBC

[Ashish K. <ash...@gm...>]

Hi,

I am a new user of openSBC product.
I was just going through it and have a quiestion.

Why it is  necessary to parse RTP. Why cannot we use only the IP and port in
the source and destination fiels to route the messages. why such type of
details are needed to be known by
ssrc=490290042 ntp=2007/8/8-2:22:0.995224 rtp=76800 psent=2 osent=320
ssrc=490290042 fraction=0 lost=0 last_seq=131074 jitter=0
lsr=419:09:26.072dlsr=0

[OpenSIPStack] Dtmf Detection (RFC 2833) in OpenSBC

[Gaurav K. <gkh...@is...>]

Hello,

 

I'm facing an issue with RTP on OpenSBC 1.1.4 in capturing RFC 2833 packets.
My network configuration is:-

 

Xlite UA (Outbound) (.64) <--------> OpenSBC (.115) <-----------> Xlite UA
(Inbound) (.60)

 

* Outbound UA is registered on OpenSBC

* Inbound UA is registered on OpenSBC

* A route exists from OpenSBC to Xlite UA (Inbound) (.60)

 

RTP_Session::SendReceiveStatus RTP_Session::OnReceiveData(const
RTP_DataFrame & frame) is the function where all rtp packets get processed.

 

The source file rfc2833.cxx has the function
OpalRFC2833Proto::ReceivedPacket() which seems to be there for receiving an
RFC2833 packet but I could not find a reference from where it's being
invoked. Is this a bug in OpenSBC?

 

Attached are the level 5 logs for inbound and outbound call where both UAs
tried to send & receive DTMF as RFC 2833 packets.

 

I'll be thankful for any guidance on what's going wrong.

 

Regards,
Gaurav

Re: [OpenSIPStack] Record phone call

[tomach <to...@dg...>]

Hello Joegen,

How are things going?

Re: [OpenSIPStack] [OpenSBC] OpenSBC behind NAT/router

[Joegen E. B. <joe...@gm...>]

Hi Roland,

There is no provision yet to set the contact address for opensbc to bind 
to the external NAT IP.   However most SIP implementations already 
support RFC 3581 (see http://www.opensipstack.org/rfc/rfc3581.txt ) so 
there might not be a need to change the contact uri.

Joegen


Roland Auckenthaler wrote:
> Hi Joegen,
>
> Last week i was back in the office and able to do some tests. I set the "Static RTP Media Address" to the public ip address and now the RTP seems to get streamed from the client the the public ip address. I still need to check if the NAT router will forward the RTP packets. 
>
> However i have also found that the Contact in the 200OK shows that internal IP adress rather than the public ip-address of the SBC behind the NAT. Is there a config field to "force" the SBC to pass back a specific IP address in the contact header of the 200OK?
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> Opensipstack-osbcdevel mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensipstack-osbcdevel
>
>   

Re: [OpenSIPStack] RTP Issue - opensipstack 1.1.5

[Gaurav K. <gkh...@is...>]

Joegen,

Thanks for your reply. I'm planning to upgrade to the latest version. Is
there a release note/ README which captures the changes/ differences =
between
stable releases of the stack & sbc. E.g. Difference between 1.1.5 & =
1.1.7
(latest & greatest) versions of opensipstack

Thanks in advance for your reply.

Regards,
Gaurav



> -----Original Message-----
> From: Joegen E. Baclor [mailto:joe...@gm...]
> Sent: Tuesday, August 07, 2007 11:53 AM
> To: Gaurav Kheterpal
> Cc: ope...@li...; ag...@de...
> Subject: Re: RTP Issue - opensipstack 1.1.5
>=20
> Inline,
>=20
> Gaurav Kheterpal wrote:
> >
> > Hello,
> >
> > I=92m facing an issue with RTP on OpenSBC 1.1.2. My network
> > configuration is:-
> >
> > Xlite UA (Outbound) (.64) =DF----=E0 OpenSBC (.115) =DF-------=E0 =
FreeSwitch
> > (.57) =DF------=E0 Xlite UA (Inbound) (.57)
> >
> > * Outbound UA is registered on OpenSBC
> >
> > * Inbound UA is registered on FreeSwitch
> >
> > * A route exists from OpenSBC to FreeSwitch
> >
> > I am using:-
> >
> > Opensipstack 1.1.5
> >
> > Opensbc 1.1.2
> >
> > I=92m seeing the following errors in Level 5 log files:-
> >
> > 2007/08/07 10:46:33.745 OpalMediaThread:af4eb8 Info RTP_UDP Started
> > Media Stream
> >
> > 2007/08/07 10:46:34.877 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7614 ssrc=3D41
> >
> > 2007/08/07 10:46:35.097 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7644 ssrc=3D41
> >
> > 2007/08/07 10:46:35.387 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7694 ssrc=3D41
> >
> > 2007/08/07 10:46:35.698 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7744 ssrc=3D41
> >
> > 2007/08/07 10:46:36.018 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7793 ssrc=3D41
> >
> > 2007/08/07 10:46:36.299 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7842 ssrc=3D41
> >
> > 2007/08/07 10:46:36.689 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7893 ssrc=3D41
> >
> > 2007/08/07 10:46:36.980 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7942 ssrc=3D41
> >
> > 2007/08/07 10:46:37.390 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 7992 ssrc=3D41
> >
> > 2007/08/07 10:46:37.761 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 8041 ssrc=3D41
> >
> > 2007/08/07 10:46:52.100 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 8092 ssrc=3D41
> >
> > 2007/08/07 10:46:52.400 OpalMediaThread:aed1f8 Info RTP Abnormal
> > change of sequence numbers, adjusting to expect 8141 ssrc=3D41
> >
> > 2007/08/07 10:46:53.692 Call Info *** DESTROYED *** SIPTimer Manager
> >
> > The Level 5 Log & Ethereal capture on OpenSBC is attached. I=92m =
able to
> > hear audio at both ends.
> >
> > I looked up the code & found the following in rtp.cxx:-
> >
> > // Check for Cisco bug where sequence numbers suddenly start
> incrementing
> >
> > // from a different base.
> >
> > if (++consecutiveOutOfOrderPackets > 10) {
> >
> > expectedSequenceNumber =3D (WORD)(sequenceNumber + 1);
> >
> > PTRACE(1, "RTP\tAbnormal change of sequence numbers, adjusting to
> > expect " << expectedSequenceNumber << " ssrc=3D" << syncSourceIn);
> >
> > }
> >
> > Is it a known issue or something which has been fixed in 1.1.7?
> >
> Of course an upgrade is always a good idea.
>=20
> This is not a bug at all. One cause might be RTP packets are lost
> between FreeSwitch and OpenSBC. OpenSBC is expecting the sequence =
number
> to increase by one. See excerpts from RFC3550 below
>=20
>     sequence number: 16 bits
>           The sequence number increments by one for each RTP data =
packet
>           sent, and may be used by the receiver to detect packet loss =
and
> to
>           restore packet sequence.  The initial value of the sequence
> number
>           SHOULD be random (unpredictable) to make known-plaintext =
attacks
>           on encryption more difficult, even if the source itself does =
not
>           encrypt according to the method in Section 9.1, because the
>           packets may flow through a translator that does.  Techniques =
for
>           choosing unpredictable numbers are discussed in [17].
>=20
>=20
>=20
> If the sequence received by OpenSBC is greater than what it is
> expecting, it would be spitting out the warning traces you have just
> pasted.
>=20
>=20
> > Moreover, when I press DTMF, I see it in Ethereal capture on OpenSBC
> > but OpenSBC does not seem to be detecting it. That=92s a different =
issue
> > from the one reported above.
> >
>=20
> OpenSBC relays DTMF as any normal RTP packets. It does not have to
> detect it.
> >
> > I=92ll be thankful for any guidance on what=92s going wrong.
> >
> > Regards,
> > Gaurav
> >
> > =
------------------------------------------------------------------------
> >
> > 2007/08/07 10:46:32.463	                  Proxy	Info	-->> From:
> sip:102@192.168.96.115 Target: INVITE
> sip:101@192.168.96.115:65062;method=3D"INVITE"
> > 2007/08/07 10:46:32.834	              UserAgent	Info	Using remote
> XML-RPC registrar at http://localhost:5678/RPC2
> > 2007/08/07 10:46:32.934	    HTTP Service:ad7198	Info
> 	Registrar.FindRegistration
> > 2007/08/07 10:46:32.954	              UserAgent	Info	No
> Registration found
> > 2007/08/07 10:46:33.745	 OpalMediaThread:aed1f8	Info	RTP_UDP
> 	Started Media Stream
> > 2007/08/07 10:46:33.745	 OpalMediaThread:af4eb8	Info	RTP_UDP
> 	Started Media Stream
> > 2007/08/07 10:46:34.877	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7614 ssrc=3D41
> > 2007/08/07 10:46:35.097	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7644 ssrc=3D41
> > 2007/08/07 10:46:35.387	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7694 ssrc=3D41
> > 2007/08/07 10:46:35.698	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7744 ssrc=3D41
> > 2007/08/07 10:46:36.018	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7793 ssrc=3D41
> > 2007/08/07 10:46:36.299	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7842 ssrc=3D41
> > 2007/08/07 10:46:36.689	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7893 ssrc=3D41
> > 2007/08/07 10:46:36.980	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7942 ssrc=3D41
> > 2007/08/07 10:46:37.390	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 7992 ssrc=3D41
> > 2007/08/07 10:46:37.761	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 8041 ssrc=3D41
> > 2007/08/07 10:46:52.100	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 8092 ssrc=3D41
> > 2007/08/07 10:46:52.400	 OpalMediaThread:aed1f8	Info	RTP
Abnormal
> change of sequence numbers, adjusting to expect 8141 ssrc=3D41
> > 2007/08/07 10:46:53.692	                   Call	Info	***
DESTROYED
> *** SIPTimer Manager
> > 2007/08/07 10:46:53.702	 OpalMediaThread:af4eb8	Info	RTP_UDP
> 	Closing Media Stream
> > 2007/08/07 10:46:53.702	 OpalMediaThread:aed1f8	Info	RTP_UDP
> 	Closing Media Stream
> >

Re: [OpenSIPStack] RTP Issue - opensipstack 1.1.5

[Joegen E. B. <joe...@gm...>]

Inline,

Gaurav Kheterpal wrote:
>
> Hello,
>
> I=92m facing an issue with RTP on OpenSBC 1.1.2. My network=20
> configuration is:-
>
> Xlite UA (Outbound) (.64) =DF----=E0 OpenSBC (.115) =DF-------=E0 FreeS=
witch=20
> (.57) =DF------=E0 Xlite UA (Inbound) (.57)
>
> * Outbound UA is registered on OpenSBC
>
> * Inbound UA is registered on FreeSwitch
>
> * A route exists from OpenSBC to FreeSwitch
>
> I am using:-
>
> Opensipstack 1.1.5
>
> Opensbc 1.1.2
>
> I=92m seeing the following errors in Level 5 log files:-
>
> 2007/08/07 10:46:33.745 OpalMediaThread:af4eb8 Info RTP_UDP Started=20
> Media Stream
>
> 2007/08/07 10:46:34.877 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7614 ssrc=3D41
>
> 2007/08/07 10:46:35.097 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7644 ssrc=3D41
>
> 2007/08/07 10:46:35.387 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7694 ssrc=3D41
>
> 2007/08/07 10:46:35.698 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7744 ssrc=3D41
>
> 2007/08/07 10:46:36.018 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7793 ssrc=3D41
>
> 2007/08/07 10:46:36.299 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7842 ssrc=3D41
>
> 2007/08/07 10:46:36.689 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7893 ssrc=3D41
>
> 2007/08/07 10:46:36.980 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7942 ssrc=3D41
>
> 2007/08/07 10:46:37.390 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 7992 ssrc=3D41
>
> 2007/08/07 10:46:37.761 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 8041 ssrc=3D41
>
> 2007/08/07 10:46:52.100 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 8092 ssrc=3D41
>
> 2007/08/07 10:46:52.400 OpalMediaThread:aed1f8 Info RTP Abnormal=20
> change of sequence numbers, adjusting to expect 8141 ssrc=3D41
>
> 2007/08/07 10:46:53.692 Call Info *** DESTROYED *** SIPTimer Manager
>
> The Level 5 Log & Ethereal capture on OpenSBC is attached. I=92m able t=
o=20
> hear audio at both ends.
>
> I looked up the code & found the following in rtp.cxx:-
>
> // Check for Cisco bug where sequence numbers suddenly start incrementi=
ng
>
> // from a different base.
>
> if (++consecutiveOutOfOrderPackets > 10) {
>
> expectedSequenceNumber =3D (WORD)(sequenceNumber + 1);
>
> PTRACE(1, "RTP\tAbnormal change of sequence numbers, adjusting to=20
> expect " << expectedSequenceNumber << " ssrc=3D" << syncSourceIn);
>
> }
>
> Is it a known issue or something which has been fixed in 1.1.7?
>
Of course an upgrade is always a good idea.

This is not a bug at all. One cause might be RTP packets are lost=20
between FreeSwitch and OpenSBC. OpenSBC is expecting the sequence number =

to increase by one. See excerpts from RFC3550 below

    sequence number: 16 bits
          The sequence number increments by one for each RTP data packet
          sent, and may be used by the receiver to detect packet loss and=
 to
          restore packet sequence.  The initial value of the sequence num=
ber
          SHOULD be random (unpredictable) to make known-plaintext attack=
s
          on encryption more difficult, even if the source itself does no=
t
          encrypt according to the method in Section 9.1, because the
          packets may flow through a translator that does.  Techniques fo=
r
          choosing unpredictable numbers are discussed in [17].



If the sequence received by OpenSBC is greater than what it is=20
expecting, it would be spitting out the warning traces you have just past=
ed.


> Moreover, when I press DTMF, I see it in Ethereal capture on OpenSBC=20
> but OpenSBC does not seem to be detecting it. That=92s a different issu=
e=20
> from the one reported above.
>

OpenSBC relays DTMF as any normal RTP packets. It does not have to=20
detect it.
>
> I=92ll be thankful for any guidance on what=92s going wrong.
>
> Regards,
> Gaurav
>
> -----------------------------------------------------------------------=
-
>
> 2007/08/07 10:46:32.463	                  Proxy	Info	-->> From: sip:102=
@192.168.96.115 Target: INVITE sip:101@192.168.96.115:65062;method=3D"INV=
ITE"
> 2007/08/07 10:46:32.834	              UserAgent	Info	Using remote XML-R=
PC registrar at http://localhost:5678/RPC2
> 2007/08/07 10:46:32.934	    HTTP Service:ad7198	Info	Registrar.FindRegi=
stration
> 2007/08/07 10:46:32.954	              UserAgent	Info	No Registration fo=
und
> 2007/08/07 10:46:33.745	 OpalMediaThread:aed1f8	Info	RTP_UDP	Started Me=
dia Stream
> 2007/08/07 10:46:33.745	 OpalMediaThread:af4eb8	Info	RTP_UDP	Started Me=
dia Stream
> 2007/08/07 10:46:34.877	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7614 ssrc=3D41
> 2007/08/07 10:46:35.097	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7644 ssrc=3D41
> 2007/08/07 10:46:35.387	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7694 ssrc=3D41
> 2007/08/07 10:46:35.698	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7744 ssrc=3D41
> 2007/08/07 10:46:36.018	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7793 ssrc=3D41
> 2007/08/07 10:46:36.299	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7842 ssrc=3D41
> 2007/08/07 10:46:36.689	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7893 ssrc=3D41
> 2007/08/07 10:46:36.980	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7942 ssrc=3D41
> 2007/08/07 10:46:37.390	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 7992 ssrc=3D41
> 2007/08/07 10:46:37.761	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 8041 ssrc=3D41
> 2007/08/07 10:46:52.100	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 8092 ssrc=3D41
> 2007/08/07 10:46:52.400	 OpalMediaThread:aed1f8	Info	RTP	Abnormal chang=
e of sequence numbers, adjusting to expect 8141 ssrc=3D41
> 2007/08/07 10:46:53.692	                   Call	Info	*** DESTROYED *** =
SIPTimer Manager
> 2007/08/07 10:46:53.702	 OpalMediaThread:af4eb8	Info	RTP_UDP	Closing Me=
dia Stream
> 2007/08/07 10:46:53.702	 OpalMediaThread:aed1f8	Info	RTP_UDP	Closing Me=
dia Stream
>  =20

[OpenSIPStack] RTP Issue - opensipstack 1.1.5

[Gaurav K. <gkh...@is...>]

Hello,

 

I'm facing an issue with RTP on OpenSBC 1.1.2. My network configuration is:-

 

Xlite UA (Outbound) (.64) <--------> OpenSBC (.115) <-----------> FreeSwitch
(.57) <----------> Xlite UA (Inbound) (.57)

 

* Outbound UA is registered on OpenSBC

* Inbound UA is registered on FreeSwitch

* A route exists from OpenSBC to FreeSwitch

 

I am using:-

 

Opensipstack 1.1.5

Opensbc 1.1.2

 

I'm seeing the following errors in Level 5 log files:-

 

2007/08/07 10:46:33.745           OpalMediaThread:af4eb8           Info
RTP_UDP           Started Media Stream

2007/08/07 10:46:34.877           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7614
ssrc=41

2007/08/07 10:46:35.097           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7644
ssrc=41

2007/08/07 10:46:35.387           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7694
ssrc=41

2007/08/07 10:46:35.698           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7744
ssrc=41

2007/08/07 10:46:36.018           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7793
ssrc=41

2007/08/07 10:46:36.299           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7842
ssrc=41

2007/08/07 10:46:36.689           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7893
ssrc=41

2007/08/07 10:46:36.980           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7942
ssrc=41

2007/08/07 10:46:37.390           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 7992
ssrc=41

2007/08/07 10:46:37.761           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 8041
ssrc=41

2007/08/07 10:46:52.100           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 8092
ssrc=41

2007/08/07 10:46:52.400           OpalMediaThread:aed1f8           Info
RTP       Abnormal change of sequence numbers, adjusting to expect 8141
ssrc=41

2007/08/07 10:46:53.692                              Call            Info
*** DESTROYED *** SIPTimer Manager

 

The Level 5 Log & Ethereal capture on OpenSBC is attached. I'm able to hear
audio at both ends.

 

I looked up the code & found the following in rtp.cxx:-

 

      // Check for Cisco bug where sequence numbers suddenly start
incrementing

      // from a different base.

      if (++consecutiveOutOfOrderPackets > 10) {

        expectedSequenceNumber = (WORD)(sequenceNumber + 1);

                  PTRACE(1, "RTP\tAbnormal change of sequence numbers,
adjusting to expect " << expectedSequenceNumber << " ssrc=" <<
syncSourceIn);

      }

 

Is it a known issue or something which has been fixed in 1.1.7? Moreover,
when I press DTMF, I see it in Ethereal capture on OpenSBC but OpenSBC does
not seem to be detecting it. That's a different issue from the one reported
above.

 

I'll be thankful for any guidance on what's going wrong.

 

Regards,
Gaurav

Re: [OpenSIPStack] SIP NOTIFY Failing

[Joegen E. B. <joe...@gm...>]

Please open a bug report in bugs.opensourcesip.org and upload the 
complete log file together with the symptoms of the problem.   Thanks.


Ryan Yaldor wrote:
> Testing OpenSBC against our Sylantro platform and registration takes place
> with no problems, but once registered, Sylantro sends out a NOTIFY message
> and that message never makes it through and eventually causes OpenSBC to
> stop responding.  Log shows a constant loop of TRYING and OK's until the
> process stops responding.
>
>  
>
> Thanks,
>
>  
>
> Ryan Yaldor
>
> PBX-Change 
>
> TampaBay DSL
>
> ph  813-243-8850 ext. 206
>
> fax 813-249-8414
>
>  <http://www.PBX-Change.com> www.PBX-Change.com
>
> www.TampaBayDSL.com <http://www.tampabaydsl.com/> 
>
>  
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> opensipstack-devel mailing list
> ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensipstack-devel
>
>
>   

[OpenSIPStack] SIP NOTIFY Failing

[Ryan Y. <ry...@ta...>]

Testing OpenSBC against our Sylantro platform and registration takes place
with no problems, but once registered, Sylantro sends out a NOTIFY message
and that message never makes it through and eventually causes OpenSBC to
stop responding.  Log shows a constant loop of TRYING and OK's until the
process stops responding.

 

Thanks,

 

Ryan Yaldor

PBX-Change 

TampaBay DSL

ph  813-243-8850 ext. 206

fax 813-249-8414

 <http://www.PBX-Change.com> www.PBX-Change.com

www.TampaBayDSL.com <http://www.tampabaydsl.com/> 

 

Re: [OpenSIPStack] FW: Proxy - CANCEL Processing

[Joegen E. B. <joe...@gm...>]

Hi gustavo,

Thanks for all these patches. I will definitely going to look at them as =

soon as I get time. I might have some free time late next week.

Gustavo Curetti wrote:
> Hi Joegen:
>
> I added code to respond 200 ok to the CANCEL like the RFC3261 said:
> **
> *
>
> 16.10 CANCEL Processing
>
> **If a matching response context is found, the element MUST=20
> immediately return a 200 (OK) response to the CANCEL request*.
>
> In ProxySession::ValidateRequest():
>
> if(m_SessionManager.GetUserAgent().GetStack().CancelInviteServerTransac=
tion(m_Request))
> {
> SIPMessage ok;
> request.CreateResponse( ok, SIPMessage::Code200_Ok );
> SendRequest( ok );
> }
>
> Please check this if you have time.
>
> Thanks for your help
>
> Gustavo
>
>
>
>     -------------------------------------------------------------------=
-----
>     From: cur...@ho...
>     To: ope...@li...
>     Subject: FW: Proxy - CANCEL Processing
>     Date: Thu, 2 Aug 2007 16:42:15 +0200
>
>         Resending with compressed logs.
>
>
>             -----------------------------------------------------------=
-------------
>             From: cur...@ho...
>             To: ope...@li...;
>             jb...@so...; joe...@gm...
>             Subject: Proxy - CANCEL Processing
>             Date: Wed, 1 Aug 2007 22:32:29 +0200
>
>             Hi Joegen:
>
>             I was checking the cancel processing in Proxy Only Mode.
>
>             From the RFC 3261:
>
>             **
>
>             *16.10 CANCEL Processing*
>
>             A stateful proxy MAY generate a CANCEL to any other
>             request it has generated at any time (subject to receiving
>             a provisional response to that request as described in
>             section 9.1). A proxy MUST cancel any pending client
>             transactions associated with a response context when it
>             receives a matching CANCEL request.
>
>             A stateful proxy MAY generate CANCEL requests for pending
>             INVITE client transactions based on the period specified
>             in the INVITE=92s Expires header field elapsing. However,
>             this is generally unnecessary since the endpoints involved
>             will take care of signaling the end of the transaction.
>
>             While a CANCEL request is handled in a stateful proxy by
>             its own server transaction, a new response context is not
>             created for it. Instead, the proxy layer searches its
>             existing response contexts for the server transaction
>             handling the request associated with this CANCEL. *_If a
>             matching response context is found, the element _**_MUST
>             immediately return a 200 (OK) response to the CANCEL
>             _**_request_*. In this case, the element is acting as a
>             user agent server as defined in Section 8.2. Furthermore,
>             the element MUST generate CANCEL requests for all pending
>             client transactions in the context as described in Section
>             16.7 step 10.
>
>             If a response context is not found, the element does not
>             have any knowledge of the request to apply the CANCEL to.
>             It MUST statelessly forward the CANCEL request (it may
>             have statelessly forwarded the associated request previousl=
y).
>
>             From the RFC 3665:
>
>             3.8. Unsuccessful No Answer
>             Alice Proxy 1 Proxy 2 Bob
>             | | | |
>             | INVITE F1 | | |
>             |--------------->| INVITE F2 | |
>             | 100 F3 |--------------->| INVITE F4 |
>             |<---------------| 100 F5 |--------------->|
>             | |<---------------| |
>             | | | 180 F6 |
>             | | 180 F7 |<---------------|
>             | 180 F8 |<---------------| |
>             |<---------------| | |
>             | CANCEL F9 | | |
>             |--------------->| | |
>             | 200 F10 | | |
>             |<---------------| CANCEL F11 | |
>             | |--------------->| |
>             | | 200 F12 | |
>             | |<---------------| |
>             | | | CANCEL F13 |
>             | | |--------------->|
>             | | | 200 F14 |
>             | | |<---------------|
>             | | | 487 F15 |
>             | | |<---------------|
>             | | | ACK F16 |
>             | | 487 F17 |--------------->|
>             | |<---------------| |
>             | | ACK F18 | |
>             | 487 F19 |--------------->| |
>             |<---------------| | |
>             | ACK F20 | | |
>             |--------------->| | |
>             | | | |
>             In this scenario, Alice gives up on the call before Bob
>             answers (sends a 200 OK response). Alice sends a CANCEL
>             (F9) since no final response had been received from Bob.
>             If a 200 OK to the INVITE had crossed with the CANCEL,
>             Alice would have sent an ACK then a BYE to Bob in order to
>             properly terminate the call.
>
>             _Note that the CANCEL message is acknowledged with a 200
>             OK on a hop by hop basis, rather than end to end._
>
>             But the OpenSBC behavior in Proxy Mode is:
>
>             Alice Proxy-OpenSBC Bob
>             | INVITE | |
>             |--------------->| INVITE |
>             | 100 |--------------->|
>             |<---------------| 180 |
>             | 180 |<---------------|
>             |<---------------| |
>             | CANCEL | |
>             |--------------->| |
>             | 487 | |
>             |<---------------| CANCEL |
>             | ACK |--------------->|
>             |--------------->| 200 |
>             | |<---------------|
>             | | 487 |
>             | |<---------------|
>             | | ACK |
>             | 200 |--------------->|
>             |<---------------| |
>
>
>
>             I attached a log (CANCEL-OK.log) and a sip flow
>             (CANCEL-OK.jpg).
>
>             I have two problems with the cancel:
>
>             -When the callee don't answer, the caller retransmit the
>             cancel, because the OpenSBC don't send the 200 ok response
>             for the cancel. Log: CANCEL-RETRANS.log , Sip Flow:
>             CANCEL-RETRANS.jpg
>
>             -When the callee send the 487(Request Cancelled) response
>             for the INVITE before the 200 ok response for the CANCEL,
>             the 200 ok is not been proxy by the OpenSBC. Log:
>             CANCEL-NOTOK.log , Sip Flow: CANCEL-NOTOK.jpg. In the log,
>             the sesion is deleted by the 487:
>             :
>
>             222:22:52.761 DBG: [CID=3D0x0475] Finding transaction for
>             SIP/2.0 487 Request Terminated
>             222:22:52.762 DBG: [CID=3D0x0475] Setting Transaction ID to=

>             955903155@192.168.0.60|z9hG4bKe531e528fbc6f89e|INVITE
>             <mailto:955903155@192.168.0.60%7Cz9hG4bKe531e528fbc6f89e%7C=
INVITE>
>             222:22:52.762 DTL: [CID=3D0x0475] Found
>             IST|955903155@192.168.0.60|z9hG4bKe531e528fbc6f89e|INVITE
>             <mailto:IST%7C955903155@192.168.0.60%7Cz9hG4bKe531e528fbc6f=
89e%7CINVITE>
>             for SIP/2.0 487 Request Terminated
>             222:22:52.763 DTL: [CID=3D0x0475] IST(800569658)
>             Event(SIPMessage) - SIP/2.0 487 Request Terminated
>             222:22:52.763 DBG: [CID=3D0x0475] TRANSACTION: (IST) SIP/2.=
0
>             487 Request Terminated State: 3
>             222:22:52.764 DTL: [CID=3D0x06cb] *** QUEUED FOR DELETION
>             *** SIPSession: 955903155@192.168.0.60
>             <mailto:955903155@192.168.0.60>
>             222:22:52.764 DBG: [CID=3D0x0000] GC: First Stale Object
>             ProxySession
>             _222:22:52.765 ERR: [CID=3D0x0000] GC:
>             .\src\ProxySession.cxx:601
>             ProxySession::OnCheckRoutePolicy:: Attempt to
>             CreateReference() a garbage collected pointer_
>             _222:22:52.766 ERR: [CID=3D0x0000] GC:
>             .\src\ProxySession.cxx:625 ProxySession::OnFinalResponse::
>             Attempt to CreateReference() a garbage collected pointer_
>
>             Any idea? Thanks for your help.
>
>             Gustavo
>
>
>
>
>
>
>
>             -----------------------------------------------------------=
-------------
>             Descubre Live.com - tu propia p=E1gina de inicio,
>             personalizada para ver r=E1pidamente todo lo que te interes=
a
>             en un mismo sitio. todo en el mismo sitio.
>             <http://www.live.com/getstarted>=20
>
>
>         ---------------------------------------------------------------=
---------
>         Env=EDa mensajes de correo electr=F3nico directamente a tu blog=

>         con MSN. Carga chistes, fotograf=EDas y muchas otras cosas. Es
>         gratis.
>         <http://clk.atdmt.com/MSN/go/msnnksac0030000001msn/direct/01/?h=
ref=3Dhttp://www.imagine-msn.com/spaces>
>
>
>
>     -------------------------------------------------------------------=
-----
>     Comun=EDcate al instante con Windows Live Messenger Windows Live
>     Messenger
>     <http://imagine-msn.com/messenger/launch80/default.aspx?locale=3Des=
-ar&source=3Djoinmsncom/messenger>
>
>
>
> -----------------------------------------------------------------------=
-
> Se uno de los primeros en probar Windows Live Mail. Windows Live Mail. =

> <http://ideas.live.com/programpage.aspx?versionId=3D5d21c51a-b161-4314-=
9b0e-4911fb2b2e6d>