Users
the nat in Ser is managed by the SIP message, i was wondering if is theer a
solution for managing the NAT in RTP protocol as the RTP is a trasport layer
protocol
All the best
Gustavo Villegas
Hi everybody, Im trying to compile rtpproxy in SOLARIS 9(SPARC), but Im
getting errors related to sys/cdefs.h and err.h, i have readen many articles
about it, but there is nothing helpful.
If anybody can help me, Ill thank it.
Atentamente!
Alejandro Mendoza
Ing. de Soporte
EPSSILON, Guatemala
Tel: (502) 22560068
Móvil: (502) 57193432
I ran into exactly the same problem in December. There are several
reasons why SER may not realize the ACK belongs to the transaction.
Look carefully at the via: and from: headers in the INVITE and the ACK.
The via: should have a branch parameter, and it should be the same in
both. Some UAs do not put in the branch parameter. In this case SER
should (according to RFC3261) be looking at the combination of the
Request URI, To tag, From tag, Call-ID, Cseq, and top via header. In my
case the problem was that SER compares the From field rather than the
>From tag, and it does so incorrectly. I raised this issue both here and
in the serdev list, (you can find the thread in the December archives),
but the developers said they would not fix this because it is in support
of an obsolete spec (RFC 2543).
I recommend you check those headers to see if your situation actually
matches what I ran into, and if so, raise the issue to the serdev list.
If enough people complain, they might decide to provide better support
of RFC3261, section 17.2.3 (Matching Requests to Server Transactions).
Dave
_____
From: serusers-bounces(a)iptel.org [mailto:serusers-bounces@lists.iptel.org] On
Behalf Of PHAN, Quang-Minh (Docteur Ordinateur)
Sent: Thursday, January 06, 2005 12:10 PM
To: serusers(a)lists.iptel.org
Subject: [Serusers] Problem with ACK on Negative Reply
Hi all,
When I use SER in transaction stateful mode, I have the following
problem:
When A sends an "INVITE B" to SER, SER forwards the INVITE to B
statefully. If B reply with "200 OK", everything goes fine as SER will
delete the transaction from memory when it sees the 200 OK. However, if
B send an "BUSY HERE" back to SER, SER will send an ACK to B right away
and then forwards the "BUSY HERE" to A. A will then send an ACK to SER
but SER won't realize that the ACK belong to the transaction and forward
it to B. SER will then keep sending "BUSY HERE" to A until the
transaction is timed out.
I wonder if there is something to add in the configuration file to
make SER understand that the ACK send from A should terminate the
transaction.
Any help will be highly appreciated!
Minh
At 07:09 PM 1/6/2005, PHAN, Quang-Minh (Docteur Ordinateur) wrote:
>Hi all,
>
> When I use SER in transaction stateful mode, I have the following problem:
>
> When A sends an "INVITE B" to SER, SER forwards the INVITE to B statefully. If B reply with "200 OK", everything goes fine as SER will delete the transaction from memory when it sees the 200 OK. However, if B send an "BUSY HERE" back to SER, SER will send an ACK to B right away and then forwards the "BUSY HERE" to A. A will then send an ACK to SER but SER won't realize that the ACK belong to the transaction and forward it to B.
It will.
you need to make sure that the ACK hits t_relay, which will consume it.
-jiri
> SER will then keep sending "BUSY HERE" to A until the transaction is timed out.
>
> I wonder if there is something to add in the configuration file to make SER understand that the ACK send from A should terminate the transaction.
>
>Any help will be highly appreciated!
>
>Minh
>_______________________________________________
>Serusers mailing list
>serusers(a)lists.iptel.org
>http://lists.iptel.org/mailman/listinfo/serusers
--
Jiri Kuthan http://iptel.org/~jiri/
Hi all,
When I use SER in transaction stateful mode, I have the following
problem:
When A sends an "INVITE B" to SER, SER forwards the INVITE to B
statefully. If B reply with "200 OK", everything goes fine as SER will
delete the transaction from memory when it sees the 200 OK. However, if B
send an "BUSY HERE" back to SER, SER will send an ACK to B right away and
then forwards the "BUSY HERE" to A. A will then send an ACK to SER but SER
won't realize that the ACK belong to the transaction and forward it to B.
SER will then keep sending "BUSY HERE" to A until the transaction is timed
out.
I wonder if there is something to add in the configuration file to make
SER understand that the ACK send from A should terminate the transaction.
Any help will be highly appreciated!
Minh
Hi,
I'm running SER on dual stack (IPv4 & IPv6) linux box and SER is listening
on both IPv4 and IPv6 ports.
Now I want to try SER's connectivity on IPv6 and seeking IPv6 SIP phone
(softphone).
Does anyone know IPv6 compatible softphone? (preferably free stuff)
--
Kuniyoshi Murata.........................iChat/AIM:macwebcaster
English-Japanese Interpreter mailto:kuni@ej-interpreter.net
Macintosh Webcast Specialist http://www.macwebcaster.com
My ISP, which holds both the broadband and landline telephony monopolies
in this area, blocks port 5060. Also, I am behind NAT (at this location;
my router has a real address).
I am hoping it would be possible to set up a proxy at a 3rd location to
make it possible to use the Broadvoice service. I am able to use FWD for
incoming and outgoing without trouble, so I don't think there are any
other obstacles.
I tried setting up my 3rd-location server to simply reflect packets that
hit it on UDP port 7000, so they went to Broadvoice's server on port 5060.
Unsurprisingly, I suppose, that didn't work. Tcpdump showed that nothing
ever came back along that route; I am not sure whether Broadvoice's server
was sending replies anywhere, but they sure weren't making it to my
intermediate reflecting machine.
So I guess there needs to be some rewriting going on. It seems like SER
may be the tool for this, but after looking through many example
configurations, I was unable to find something as simple as what I'm
after. Does anyone have any ideas about a good place for me to start? Is
SER in fact an appropriate tool for this task?
Thanks.
miguel
This is taken from a older answer from me.
Hello.
As far as i know there is a new command called max_contacts(number),
where "number" is the maximun number of contacts that is possible to
register for a single user. Unfortunately, this command is available only
in the new 0.9 version.
I tested it and it seems to do the trick ok.
Cheers
Ricardo Martinez
-----Mensaje original-----
De: Martin Bangiev [mailto:martin.bangiev@borsabg.com]
Enviado el: Jueves, 06 de Enero de 2005 10:31
Para: serusers(a)lists.iptel.org
Asunto: [Serusers] REGISTER
Hello people,
I'm trying to forbid the SIP clients to have more than one registered
contact at a time, but with no success. Does anyone has an idea how can
this be done? I want to keep only the last registered location.
Thanks in advance
Martin
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
Hello people,
I'm trying to forbid the SIP clients to have more than one registered
contact at a time, but with no success. Does anyone has an idea how can
this be done? I want to keep only the last registered location.
Thanks in advance
Martin
Hello,
My company is looking for a SER consultant in Montreal, QC, CAN to assist
with implementing a SER/Asterisk solution. This will involve configuration
and most likely installation. Please contact me off list if you are or
someone you know is interested.
Thank you,
David Filion
Email: dfilion(a)dotality.com