2008/1/8, Fabio Margarido <[EMAIL PROTECTED]>:
>I suspect it is the same
> issue I described in my other thread
> (http://sourceforge.net/mailarchive/forum.php?thread_name=1715019e0712120241j71323d0bs6ae65bd9bd60166c%40mail.gmail.com&forum_name=sofia-sip-devel),
> in which Mikhail Zabaluev suggeste
2008/1/21, Bernhard Suttner <[EMAIL PROTECTED]>:
> I think the best solution is as follows:
>
> - remove SOATAG_HOLD: Ok, I know that this is not so simple, therefore
> we should think about marking it only as deprecated but this will
> require more work for the implemenation.
> - create a new SOA
2008/1/18, Jerry Richards <[EMAIL PROTECTED]>:
> Another idea you might wish to consider: Regarding amplification of "900
> Internal NUA Error" logging: What I did locally to the
> nua_client_init_request() function of libsofia-sip-ua/nua/nua_stack.c is to
> set a local variable before each "goto
2008/1/18, Jerry Richards <[EMAIL PROTECTED]>:
> If I add the SIPTAG_P_PREFERRED_IDENTITY() tag to my nua_invite(), then I
> get a "900 NUA Internal Error" generated after the second call to
> sip_add_tagis() function in the nua_client_init_request() function of
> libsofia-sip-ua/nua/nua_stack.c.
2008/1/18, Martin Drasar <[EMAIL PROTECTED]>:
> I have a problem with proxy authentication.
>
> My user agent registers itself with proxy without a problem. However
> when I want it to redirect a call, the following happens:
>
> 1) REFER is sent to proxy
> 2) Proxy answers with 407
> 3) In nua_r_re
, or
> sendonly or something else.
>
> Is it possible to set a option to sofia sip that i want to send the 200
> OK (5.) with sendrecv and not with inactive?
I guess the soa_sdp_mode_set_is_needed() has to be fixed so that it
uses the user SDP and SOATAG_HOLD() to determine the wanted
2008/1/17, Jerry Richards <[EMAIL PROTECTED]>:
> Am I the only one who finds debugging a "900 Internal NUA Error" difficult
> to isolate? After a unit is deployed to a customer, if they get "900
> Internal NUA Error", I think it will be an order magnitude more difficult to
> debug, without amplifi
2008/1/18, Bernhard Suttner <[EMAIL PROTECTED]>:
> ok I have tested the case where the remote send a inactive to sofia sip
> and it does work. Also the case where sofia-sip send a inactive (hold=#)
> or a sendonly (hold=*) does work.
Cool.
> Do you know when the 1.12.8 will be released?
If there
stag, den 17.01.2008, 14:37 +0200 schrieb Pekka Pessi:
> > 2008/1/16, Bernhard Suttner <[EMAIL PROTECTED]>:
> > > I think I could fix that bug but don't know how
> > > (ok, I think the problem is in soa/soa_static.c function
> > > soa_sdp_mode_set).
> &g
Hello all,
I've put a tarball for first 1.12.8 pre-release available from
http://sofia-sip.org/~ppessi/sofia-sip-1.12.7pre1.12.8.1.tar.gz
The pre-release corresponds to the current darcs version from
http://sofia-sip.org/repos/sofia-sip/
Please drop a note if you find something fatal with the
2008/1/16, Mario Bensi <[EMAIL PROTECTED]>:
> I try to use sofsip-cli to test sofia-sip on linux and I have this error when
> i try to call a classic phone :
>
> UA: INVITE: 488 No vocodeur intersection
The endpoints don't have common speech codecs.
> My registart is sip:freephonie.net
> Do you
2008/1/16, Bernhard Suttner <[EMAIL PROTECTED]>:
> I think I could fix that bug but don't know how
> (ok, I think the problem is in soa/soa_static.c function soa_sdp_mode_set).
>
> Can somebody give me a hint how to fix that because I dont know what the
> "sm-> and rm-> in the
> soa_sdp_mode_set f
2008/1/14, Magnus Correa Marques Russo <[EMAIL PROTECTED]>:
> I've found this issue when I was running in linux and doing a lot of calls
> during overnight.
> I don't know when exactly this happens, but it just happens under high
> traffic.
Looks like dangling references or corrupted memory... Can
the forking, I think Misha also
needs the fix...
--Pekka
> Am Freitag, den 21.12.2007, 15:06 +0200 schrieb Pekka Pessi:
> > 2007/12/14, Bernhard Suttner <[EMAIL PROTECTED]>:
> > > I try to do my best and look after this bug. But I have a another
> > > question to
2008/1/8, Jerry Richards <[EMAIL PROTECTED]>:
> Do you all agree that the transferor should transmit the BYE for the first
> (held) leg in this scenario?
I think that is the right thing to do. Have a peek in
draft-ietf-sipping-service-examples.
The
http://www.tech-invite.com/Ti-sip-service-5.htm
2008/1/4, [EMAIL PROTECTED] <[EMAIL PROTECTED]>:
> Looks like I will have to get in sofia's code. However, I don't now the very
> insides of SIP and sofia-sip, so I think it will take quite a long time.
>
> On the other side, if anyone is interested in this, this error is thrown
> with sofsip-cli
2008/1/9, Francesco Lamonica <[EMAIL PROTECTED]>:
> Hi, i found a problem on win32 with the latest windows SDK (the one
> with vista support)
> in su.h (line 73 - sofia 1.12.7) it does not enter in the "then"
> branch of the #if defined(IPPROTO_IPV6) and tries to use the
> nonexistent tpipv6.h
> t
2008/1/8, Jerry Richards <[EMAIL PROTECTED]>:
> When I invoke a consultative transfer operation using nua_refer(), sofia-sip
> does not include the "Event: refer;id=93773887" field in the 200 OK reply to
> a NOTIFY.
>
> Here is the sequence beginning with two calls active (one on-hold awaiting
> tr
2007/12/6, Jan Schumacher <[EMAIL PROTECTED]>:
> has anyone implemented the feature to receive SIP Messages via SIP
> multicast socket in the Sofia-lib?
> I need something to listen on the SIP Multicast Address (RFC 3261 10.2.6).
>
> Is there anybody who has used SIP Multicast with Sofia-lib?
I du
2008/1/2, Jerry Richards <[EMAIL PROTECTED]>:
> I'm getting the following DNS error. Is there something I need to enable to
> get DNS to work?
The stack parses /etc/resolv.conf, and if there is no
/etc/resolv.conf, or it does not contain any valid nameserver entries,
sends DNS queries to localhos
2008/1/3, Fabio Margarido <[EMAIL PROTECTED]>:
> On Jan 3, 2008 11:46 AM, Pekka Pessi <[EMAIL PROTECTED]> wrote:
> > I have no idea. It seems to be that your problem is because the
> > freelist is corrupted, and it should be possible to detect that with
> > v
2008/1/2, [EMAIL PROTECTED] <[EMAIL PROTECTED]>:
> back to the point, I'm having some problems when trying to answer a call
> through wengo. I get a "strange ACK" message when accepting the invite, and
> then a terminated message with an internal error before that.
> However, it's fine if I am th
2008/1/3, Fabio Margarido <[EMAIL PROTECTED]>:
> > You can use tport tags TPTAG_UDP_RMEM() and TPTAG_UDP_WMEM() to adjust
> > the amount of buffer space. On Linux systems, you should also consider
> > adjusting rmem and wmem sysctls.
> >
> > --Pekka
>
> I'll try fiddling with those, but in the mean
2007/12/8, Jerry Richards <[EMAIL PROTECTED]>:
> I noticed if my application tries to refer an existing held call to an
> invalid number, sofia-sip will not ACK the incoming 404 Not Found response.
> I tried calling nua_ack(nh, TAG_END()) in the callback function, but this
> did not work either.
H
2007/12/13, Jerry Richards <[EMAIL PROTECTED]>:
> Does anyone know how to locally forward a call using sofia-sip? For
> example, if I get an INVITE and want it automatically forward it, what is
> the sofia-sip nua_x() function to do this?
I think you are looking for nua_respond() with status
2007/12/18, Fabio Margarido <[EMAIL PROTECTED]>:
> I'm getting a bunch of these after a large number of calls in my application:
>
> tport_vsend(0x81276a8): No buffer space available with (s=9
> UDP/10.40.0.9:5060)\par
> incoming_reply: tport_tsend: error (No buffer space available) while
> sending
2007/12/14, Bernhard Suttner <[EMAIL PROTECTED]>:
> I try to do my best and look after this bug. But I have a another
> question to the structure of sofia sip.
>
> The situation doesn't changed:
>
> NUA Application calls Alice over Broadworks PBX.
>
> The question is, in which function does NUA/NTA
2007/11/7, Martin Drasar <[EMAIL PROTECTED]>:
> Hello everyone,
> first of all - sorry if I am asking already answered question, but I
> couldn't manage to make search function on mailing list to work :o(
>
> I have troubles with Linksys SPA941 sipphone. Whenever I place an
> outbound call from my
007-12-20 at 15:13 +0200, Pekka Pessi wrote:
> > 2007/12/7, Ge@@ru <[EMAIL PROTECTED]>:
> > > I try to use STUN feature of Sofia i have the following problem:
> > >
> > > If i try to destroy stun_handle_t (with stun_handle_destroy) inside
> > >
2007/12/10, Jerry Richards <[EMAIL PROTECTED]>:
> Hello All,
>
> I have a general question.
>
> In order to fix the memory leak in 1.12.7, I have a choice. Either, I get
> the release version 1.12.7 and only apply the one "moved message passing
> into nua_stack" patch, OR I get the latest darcs ve
2007/12/7, Ge@@ru <[EMAIL PROTECTED]>:
> I try to use STUN feature of Sofia i have the following problem:
>
> If i try to destroy stun_handle_t (with stun_handle_destroy) inside
> stun_bind_cb there is always a segfault at row 2001 of stun.c file
> because sofia try to access to req variable after
Oops.
-- Forwarded message --
From: Pekka Pessi <[EMAIL PROTECTED]>
Date: 30.11.2007 16:03
Subject: Re: [Sofia-sip-devel] "Internal NUA Error" followed by crash
To: Fabio Margarido <[EMAIL PROTECTED]>
2007/11/28, Fabio Margarido <[EMAIL PROTECTED]>
2007/11/22, Fabio Margarido <[EMAIL PROTECTED]>:
> I'd like to know if there is a way for me to tell if a response my
> callback gets came from a remote host or was generated by the
> underlying stack layer. For instance, if I send an INVITE and get an
> ICMP, Sofia automatically generates a 503 re
2007/11/12, Ge@@ru <[EMAIL PROTECTED]>:
> My target:
> Use stun_bind on socket created for receive my data.
> At moment, seems that this is't possible because if i try to call
> stun_bind on a socket where i insert a su_wait_t object (for IN and ERR)
> i have an error like this:
> EPOLL_CTL_ADD(7,
2007/11/13, Fabio Margarido <[EMAIL PROTECTED]>:
> On Nov 4, 2007 7:33 PM, Pekka Pessi <[EMAIL PROTECTED]> wrote:
> > In my reading of the spec the section 13.2.2.3 refers to "UAC core"
> > layer (in Sofia, NUA) but the second final error response is a
2007/11/8, Tan Miaoqing <[EMAIL PROTECTED]>:
> However, I don't really get its meaning. Thus, I use sip_event_make method
> to construct a event header instead, and then pass it to nua_invite as
> follows:
There should be a NUTAG_REFER_EVENT() in nua_i_refer tags, the
intention is that you should
2007/11/6, Youness Alaoui <[EMAIL PROTECTED]>:
> Attached is a patch that will allow sofia-sip to compile correctly for
> the iPhone. The issue is that many files (20 files in total) are using
> printf/fprintf/etc.. without ever including stdio.h which is needed to
> use these functions. Apparently
2007/11/15, Jerry Richards <[EMAIL PROTECTED]>:
> I have attached two types of files. One file type is a text file that
> contains logs from the phone for this trace. The other files are Linux
> screenshot .png files that show the call history captured from my In-Circuit
> Debugger at the point t
Jerry,
Could you get a backtrace out from core dump (or better yet, run your
app under debugger and get backtrace from the faulting thread)?
--
Pekka.Pessi mail at nokia.com
-
This SF.net email is sponsored by: Splunk Inc.
nd
callbacks from tport). If the connection is closed by application in
normal order (and if TPTAG_SDWN_ERROR(0) is not given), the pend
callbacks should be called, too, so the application waiting for
shutdown event will catch it.
--Pekka
> On Tue, 2007-11-06 at 00:04 +0200, Pekka Pessi w
2007/11/5, Ge@@ru <[EMAIL PROTECTED]>:
> i'm using sofia 1.12.7 and i trying to use tport_pend to handled close
> of connection, but i have some problems:
>
> What i done on my sources:
>
> On tport object recover from incoming request a set this params:
> tport_set_params(tport, TPTAG_KEEPALIVE(1
2007/11/5, Bernhard Suttner <[EMAIL PROTECTED]>:
> Fine, then I think sofia-sip should support it. I will try to implement
> it. Would it be OK to set another option to SOATAG_HOLD - f.e. if you
> set SOATAG_HOLD to "#" then the connection is set to inactive?
The "*" is just s wildcard. Perhaps so
2007/11/5, [EMAIL PROTECTED] <[EMAIL PROTECTED]>:
> >Mikhail - you are correct, the INVITE (and SUBSCRIBE) can end up with
> >multiple responses from different dialogs (and UPDATE/NOTIFY requests)
> >within the same handle.
>
> Right, but how do I tell the dialogs apart in INVITE responses?
The To
2007/11/5, [EMAIL PROTECTED] <[EMAIL PROTECTED]>:
> >There can be only one ordinary client transaction going on within a
> >dialog at a time. Also, there can be only one dialog within handle.
> Wasn't true for me with 1.12.6 and outgoing INVITE -- responses from
> different dialogs in a forked ca
2007/11/1, Jerry Richards <[EMAIL PROTECTED]>:
> Is there a way for the application to obtain the value of the CSEQ field
> that sofia-sip sends in an outgoing NOTIFY message? I need this in order to
> distinguish multiple outgoing NOTIFY requests with multiple incoming 200
> OKs.
You are not usi
2007/11/2, Lu, Qichi <[EMAIL PROTECTED]>:
> I'm using 1.12.4. It looks like that the re-INVITE doesn't get passed to
> my na callback. How to enable the stack to pass the re-INVITE my
> callback so I can change my RTP session?
You have to turn off the auto-answer mode explicitly, include tag
NUTAG
2007/10/31, Fabio Margarido <[EMAIL PROTECTED]>:
> I'm observing a small issue with Sofia in which it ACKs two sequential
> non-2XX (500 and 603, in my case) responses to an outgoing INVITE
> transaction. Section 13.2.2.3 of RFC3261 says that after the first
> non-2XX response "subsequent final res
2007/10/31, Liang <[EMAIL PROTECTED]>:
> I'm trying to find out a way to log sofia sip stack's behavior, like
> logging how a SIP
> registration/publication/subcribe/bye/cancel/message
> REQUEST being sent out with what , , , ,
> SIP headers, etc.
> Example :
> I'm trying to make a lo
2007/10/29, Carmelo <[EMAIL PROTECTED]>:
> is it possible to connect to the virtual interface of the SIP server?
> I have tried with classic NUA methods, but it doesn't work :-(
> If I try to connect to the physical ip address it works, but if I change it
> with the virtual one, it doesn't work ...
2007/10/29, [EMAIL PROTECTED] <[EMAIL PROTECTED]>:
> Is there a way to retrieve the transport IP address binding currently in use
> (outbound) by the NUA stack?
> Same for the inbound transport in incoming requests?
You mean the local IP address + port pair for the socket used when sending?
Not i
2007/10/30, Massimo Cetra <[EMAIL PROTECTED]>:
> Hi list,
>
> i'm trying to solve the broken natted-devices (mostly not using stun).
>
> When a sip client connects to my application (a new channel for
> callweaver) using it's internal private address, if i use
> NTATAG_SERVER_RPORT(1) , it allows m
2007/10/30, Ge@@ru <[EMAIL PROTECTED]>:
> why on hash_resize macro (su/sofia-sip/htable.h) there isn't su_free of
> old_hash? There is a su_zalloc call for allocate a new memory area but
> then i don't see su_free of old data.
Looks like a leak. I'll fix that and see if it breaks anything...
--
2007/10/29, Stefan Leuenberger <[EMAIL PROTECTED]>:
...
> Quite a simple solution to circumvent a blacklisting algorithm would be
> to re-sort the internal DNS cache on a server timeout. A server that did
> not respond is moved to the tail of the list (kind of grey-listed). The
> next transaction w
lts[0]);
> sr->sr_results++;
> if (sr->sr_results[0]) {
>
>
> Kind regards
> Stefan
>
>
> -Original Message-
> From: Pekka Pessi [mailto:[EMAIL PROTECTED]
> Sent: Thursday, October 25, 2007 6:34 PM
> To: Stefan Leuenberger
> Subject:
2007/10/26, Jerry Richards <[EMAIL PROTECTED]>:
> So does this mean that nua_notifier() will not be fixed and eventually
> become extinct? Is the nua_i_subscribe/nua_notify/nua_r_notify scheme the
> way of the future? Will this work in my version 1.12.6work4?
They serve different purposes. The n
ver. ;-)
--Pekka
> -
> Re: [Sofia-sip-devel] NTLM authenticationFrom: Pekka Pessi <[EMAIL
> PROTECTED]> - 2006-09-25 23:12
> On 9/22/06, Nuno Centeio <[EMAIL PROTECTED]> wrote:
> > I want to use your SIP library on a SIP client but I'm havi
2007/10/24, Michael Jerris <[EMAIL PROTECTED]>:
> we have this issue already narrowed down to the patch that caused it:
>
> > Mon Oct 8 07:54:24 EDT 2007 [EMAIL PROTECTED]
> > * nua.c: nua event callback does not use nua_t object after
> nua_destroy()
> > call. sf.net bug #1803686
>
> and Pekk
2007/10/25, Tan Miaoqing <[EMAIL PROTECTED]>:
> As defined in NUA document, operation handle represents an abstract SIP
> call/session. It contains information of SIP dialog and media session... I
> suppose there should be some ways for application to access dialog
> information?
nua_handle_make_r
2007/10/24, Jerry Richards <[EMAIL PROTECTED]>:
> Any idea why I'm getting an assertion in tl_gets()? It happens upon an
> nua_i_subscription event where I invoke the following:
> enum nua_substate tSubscribeState;
> tl_gets(tags, NUTAG_SUBSTATE(tSubscribeState), TAG_END());
> Th
2007/10/22, Fabio Margarido <[EMAIL PROTECTED]>:
> However, my test case shows that when the 200 OK contains Record-Route
> headers, the packet destination and the Request-URI in the ACK sent by
> Sofia is the URI in the last Record-Route, and not the URI from the
> Contact header.
If the last Rec
2007/10/17, Fabio Margarido <[EMAIL PROTECTED]>:
> is there a way I can tell Sofia to include a payload in the error
> responses sent automatically by the nua stack?
No, not at the moment. Is the payload constant, or should it change?
SIP is really bad at propagating error responses back to the c
2007/10/17, Jerry Richards <[EMAIL PROTECTED]>:
> With the current sofia-sip implementation, I will need to create a timer and
> always wait some fixed period of time for an error response before I can
> send the INVITE. The problem with this is that any chosen time period can
> be prone to error
2007/10/2, Jerry Richards <[EMAIL PROTECTED]>:
> Is this a bug in 1.12.4?
It has a jacket and tie, so it looks like a feature..
> I'm calling nua_notifier() for a dialog subscription and sofia-sip
> immediately returns with an "nua_r_notifier 200 OK" event (and NULL sip
> message pointer), which
Mikhail Zabaluev (bug fixes)
- Fabio Margarido (bug fixes)
- Martti Mela (Open C port and extra functionality, bug fixes)
- Kai Vehmanen (fixes to win32 VC6/2005 project files)
- Pekka Pessi (everything else)
See the AUTHORS file in the distribution package.
Notes on new features
2007/10/15, Tan Miaoqing <[EMAIL PROTECTED]>:
> However, I did not find this sample code in directory nua. Thus, can anybody
> tell me where I can find this nua_cli.c?
The sofsip_cli package mentioned by Marcus has replaced the nua_cli.c.
The reference to the latter is a leftover from past.
--
P
2007/10/16, Lu, Qichi <[EMAIL PROTECTED]>:
> I need to originate calls to two agents and bridge them later on. When I
> tried to use nua_invite() without SDP I got seg fault.
How did you try to do that? Which Sofia SIP version did you use?
> Anyone knows how to achieve this?
Disable soa at call
2007/10/12, Jerry Richards <[EMAIL PROTECTED]>:
> In SIP, does anyone know how an endpoint can detect that the remote endpoint
> was lost (e.g. ethernet cable disconnect and no BYE sent)? Is there an RFC
> or something on this?
The session timer extension (RFC 4028). Please see documentation for
Hello,
I've uploaded a prerelease tar package to
http://sofia-sip.org/~ppessi/sofia-sip-1.12.6work3.tar.gz
Please give it a try.
If there is nothing fatal, I'll release the 1.12.7 shortly.
--
Pekka.Pessi mail at nokia.com
RELEASE
Description: Binary data
--
2007/10/1, Stefan Leuenberger <[EMAIL PROTECTED]>:
> I can think of two possible solutions for this problem:
>
> 1. sent the INVITE to the phone-application only on nua_i_state
> (instead of nua_i_invite). Currently I am reluctant to do so since the
> documentation of nua_i_invite states that
2007/10/10, Pekka Pessi <[EMAIL PROTECTED]>:
>(Or you can just not to zero susa.su_port before bind in
> tport_base_connect() in tport.c, and see what happens)..
Looks like this works on my linux box. Whee.
--
Pekka.Pessi mail
2007/10/8, Ge@@ru <[EMAIL PROTECTED]>:
> i have some questions about use of outbound mode.
> Usually when an user agent send a register message open an ephemeral
> port and send message to port 5060. With sofia-sip library it is
> possible use port 5060 (as source port) to send message?
As I ment
2007/10/9, Jerry Richards <[EMAIL PROTECTED]>:
> What is the proper method in the NUA callback function to check whether a
> SIP message field is valid. In some cases, if my callback function tries to
> access a SIP message field that is not present, it can get a segmentation
> fault.
A non-null
2007/10/8, Nico dHeureuse <[EMAIL PROTECTED]>:
> I have exaclty the same problem described by Fabio - even with a fresh
> checkout from the darcs repositry just a few minutes ago.
> Has this problem already been solved and I just missed the solution or is it
> still an open issue?
It should be f
There is no authentication supported there.
I guess it would be relatively easy to Basic authentication if you had
username/password in the proxy URI, like TPTAG_HTTP_CONNECT("
Http://user:[EMAIL PROTECTED]:8080"). Have a peek in
libsofia-sip-ua/tport/tport_type_connect.c. You need to add a
Proxy
2007/9/25, Jerry Richards <[EMAIL PROTECTED]>:
> Could anyone provide me a patch for 1.12.6 so that I can preserve my
> application as shown below? That is, I invoke nua_create/nua_destroy from
> my main application thread, and call su_root_run() from a separate thread,
> as shown below.
I just p
2007/10/1, Fabio Margarido <[EMAIL PROTECTED]>:
> Have you seen the bug I filed in the tracker for this PRACK issue?
> Apparently there is some code missing in nua_stack.c/.h . Changing
> those files made it work, no need to manually setting RAck.
Oops. Thanks.
Looks like an unusually bad "it is
Oops.
-- Forwarded message --
From: Pekka Pessi <[EMAIL PROTECTED]>
Date: 8.10.2007 16:28
Subject: Re: [Sofia-sip-devel] Application Timers
To: Jerry Richards <[EMAIL PROTECTED]>
2007/10/2, Jerry Richards <[EMAIL PROTECTED]>:
> Is there a list of NUA fu
2007/9/27, [EMAIL PROTECTED] <[EMAIL PROTECTED]>:
> when I call nta_outgoing_tcreate(), my network trace shows that
> Sofia-SIP selects TCP transport to send the message. I would like to
> understand why it does so, and I would also like to be able to enforce
> usage of UDP.
The RFC 3261 tries to
2007/9/27, Fabio Margarido <[EMAIL PROTECTED]>:
> Hi there,
>
> when my application receives a 180 with a Require header containing
> 100rel, the PRACK requests Sofia sends lacks the user part of the url
> of the From header, which causes the remote agent to respond with a
> 481.
That is interesti
2007/9/21, Fabio Margarido <[EMAIL PROTECTED]>:
> I'm using sip_request_create as follows:
>
> sip_request_create(p_home, sip_method_invite, NULL,
> URL_STRING_MAKE(address), NULL);
>
> because of a problem in my code, the string 'address' that was passed
> to the stack was like:
>
> sip:[EMAIL PRO
2007/9/19, Pekka Pessi <[EMAIL PROTECTED]>:
> The scheme is a token that should be compared in case-insensitive
> manner, but it is not. It seems to me that Freeswitch uses Digest
> instead of DIGEST when it provides the username and pasword to the
> stack.
Hmph, looks like fs
2007/9/18, Martin Drasar <[EMAIL PROTECTED]>:
> So I have finally found where was the catch...
>
> There is a type incompatibility between sip_content_length_s->l_length
> (uint32_t) and msg_numeric_s->x_value (unsigned long)
> After changing x_value to uint32_t, things seem to be working.
>
> What
2007/9/18, Jerry Richards <[EMAIL PROTECTED]>:
> My 1.12.4 based application always called su_root_create() and nua_create()
> first, then called su_root_run() from a newly created thread. After that,
> nua_register() is called.
>
> Did something in this area change in version 1.12.6?
Unfortunate
2007/9/18, Russell McConnachie <[EMAIL PROTECTED]>:
> I am using Freeswitch as my SIP proxy, I am trying to outbound register to
> a SIP proxy/location server. Whenever I send out a REGISTER method I get
> acknowledged back with a 401 asking me to respond to a challenge. The
> challenge contains t
2007/9/11, Martin DraĊĦar <[EMAIL PROTECTED]>:
> I would like to ask you if someone has tested (and made working) sofia
> on Solaris 9 / 64bit sparc.
>
> I have problem with nua_i_invite event.
> *sip->sip_content_type has
> c_type = 0x10069f7b9 "application/sdp", c_subtype = 0x10069f7c5 "sdp"
> but
2007/9/13, Jerry Richards <[EMAIL PROTECTED]>:
> By the way, for the problem below, I'm running the same application code
> that I ran with 1.12.4. Also, I am running with Linux version 2.4.17.
> I do notice the following error that might be related to why it's not
> sending packets? But what do
2007/9/13, Fabio Margarido <[EMAIL PROTECTED]>:
> But there's also the case in which I must include a payload in the ACK to an
> error response I may get for an outgoing INVITE. The NUTAG_AUTOACK tag
> documentation states "that ACK is always sent automatically by lower layers
> of the stack after
2007/9/12, Stefan Leuenberger <[EMAIL PROTECTED]>:
> For SIPS/TLS transport, NUA establishes a listening socket on the transport
> address specified in NUTAG_URL (IP and port) when calling nua_create().
>
> The registration request to the proxy contains this transport address and
> establishes the
2007/9/10, Jerry Richards <[EMAIL PROTECTED]>:
> Well, I figured out the problem with my nua_refer() invocation. The problem
> is the sofia-sip stack was rejecting the Refer-To header because it was not
> surrounded by <> brackets. I'm a bit surprised by this.
> I thought sofia-sip normally adds
2007/9/8, Jerry Richards <[EMAIL PROTECTED]>:
> Can anyone help me with this problem? I'm really beginning to flounder on
> this one. The scenario is as follows:
>
> 1) Receive/connect incoming call.
> 2) Put call on hold.
> 3) Send/connect secondary outgoing call.
> 4) Transfer call 1 to call 2
2007/8/31, Fabio Margarido <[EMAIL PROTECTED]>:
> ok then. I don't know if you expect to be able to do it soon, but in
> the meantime, is there a quick and dirty hack I can do to get me going
> until the final solution?
Looks like the canonization function was a bit too simple... I pushed
a fix t
2007/8/29, Aleksander Piotrowski <[EMAIL PROTECTED]>:
> i'm curious if it's possible to call my own callback function once particular
> timer timesout? in my own application, i've got private structure that
> is used to store internal state of sip call.
> i see the following message "nta: timer D
Hello all,
We plan to have a meeting on our freenode IRC channel #sofia-sip next
Monday at 17:00 Finnish time (14:00 CET, or 2 pm).
The main topics are the pending issues for the release 1.12.7 and
future work beyond it, including things like GRUUs, Outbound, TLS
support, etc. We would also welco
Oops.
-- Forwarded message --
From: Pekka Pessi <[EMAIL PROTECTED]>
Date: 23.8.2007 17:54
Subject: Re: [Sofia-sip-devel] Sofia doesn't send an ACK for 487 after a CANCEL
To: Fabio Margarido <[EMAIL PROTECTED]>
2007/8/22, Fabio Margarido <[EMAIL PROTECTED]&
2007/8/22, Fabio Margarido <[EMAIL PROTECTED]>:
> As you can see, the softphone adds a 'received' parameter to the Via
> header. Is this what may be causing the message to be discarded?
Nope, only the "host-port" and branch parameter are taken into account
when comparing Via.
My guess is that the
2007/7/31, alx <[EMAIL PROTECTED]>:
> I m currently developing a sip client, which is going to communicate
> with open-ims core. I'm actually looking for a way to specify the
> destination port without sending it at the end of the REGISTER message
> ( e.g. REGISTER sip:my.open-ims.org direct to por
2007/8/3, rajeev k <[EMAIL PROTECTED]>:
> To force the application to use tcp I am using following tag with
> nua_set_params. It works fine except that the tcp will use random ports not
> the specified port.
>
> NUTAG_URL("sip:*:5060;transport=tcp,udp")
>
> Now there is a requirement that the clien
2007/8/10, Chung Pak Lai <[EMAIL PROTECTED]>:
> I have implemented a SIP server. Right now, I am trying to deal with the
> problem if the SIP client didn't send a BYE and terminated the socket
> connection. What is the best way to deal with it? For TCP SIP connection, is
> there any notification to
2007/8/15, Jerry Richards <[EMAIL PROTECTED]>:
> Is there a way to force sofia-sip to periodically re-try registration after
> a 503 Service Unavailable error?
>
> I need this two cases:
>
> 1) Never registered successfully (but keeps retrying)
> 2) Was registered, but lost registra
2007/8/15, Diego Costantini <[EMAIL PROTECTED]>:
> I get an Assertion 'su_port_own_thread(self)' failed when I call
> nta_outgoing_tcreate.
You call nta_outgoing_tcreate() from different thread that you used to
call nta_agent_create()(/su_root_create(). I guess we should use
locking there, but we
401 - 500 of 854 matches
Mail list logo
