Hi, I'm using start_recording() function of nathelper module. I use
"-R" option in rtpproxy daemon in order not to record RTCP.
start_recording() is called in route and onreply_route.
After the call ends I see the following files in the recording directory:
283K 2009-07-06 13:41
1e54fbae418504a40744ad765718792a(a)212.230.19.1=38a3b929-co1782-INS001;1.a.rtp
275K 2009-07-06 13:41
1e54fbae418504a40744ad765718792a(a)212.230.19.1=38a3b929-co1782-INS001;1.o.rtp
0 2009-07-06 13:41
1e54fbae418504a40744ad765718792a(a)212.230.19.1=38a3b929-co1782-INS001;2.a.rtp
0 2009-07-06 13:41
1e54fbae418504a40744ad765718792a(a)212.230.19.1=38a3b929-co1782-INS001;2.o.rtp
Some questions:
- "1.a.rtp" is one direction and "1.o.rtp" the other, right?
- What are the last two files (0 size)?
- How to extract the audio from these files? note that the used codec
could be G729, ALAW, GSM...
Thanks a lot.
--
Iñaki Baz Castillo
<ibc(a)aliax.net>
I looked around... CR uses the same approach that LCR on prefix
definitions (or, at least the README don't say/explain that it's
different). So I went to look on REGEX module...
Iñaki, this module has Your signature... let me ask You what exactly are
this REGEX groups? Could I use something like:
- Use RURI and recover associated "REGEX-group number";
- Ask LCR or CR to look for routes to prefix="REGEX-group number";
- proceed routing fail-over as usual...
Edson
Juha Heinanen escreveu:
> Iñaki Baz Castillo writes:
>
> > I'd really like to see some mechanism to make routing decisions based
> > on RURI username itself, rather than just no its prefix. Sure it
> > could be a good feature.
>
> lcr module works on longest prefixes of r-uri username (= telephone
> number). it was tried once operate on regular expressions, but it was
> not a good idea. if you want something else than longest prefixes,
> check if some other module (like drouting or cr) could provide that.
>
> -- juha
>
> _______________________________________________
> sr-dev mailing list
> sr-dev(a)lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
>
Henning, how can I configure Kamailio to use some other path to find
modules? i should make modules differently as well, right?
I'll copy back the list so anyone can answer that.
Rgds,
Uriel
On Wed, Jul 1, 2009 at 9:35 AM, Henning Westerholt <
henning.westerholt(a)1und1.de> wrote:
> On Mittwoch, 1. Juli 2009, you wrote:
> > It's a CentOS 5.3 i386.
> >
> > # gcc --version
> > gcc (GCC) 4.1.2 20071124 (Red Hat 4.1.2-42)
> > Copyright (C) 2006 Free Software Foundation, Inc.
> > This is free software; see the source for copying conditions. There is NO
> > warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
> PURPOSE.
> >
> > # uname -a
> > Linux sbci1pri.cpsnet.com.ar 2.6.18-92.1.18.el5 #1 SMP Wed Nov 12
> 09:30:27
> > EST 2008 i686 i686 i386 GNU/Linux
>
>
> Hi Uriel,
>
>
> thanks, but this is all fairly common, so i doubt your environment is
> guilty here.
>
>
> > Do you have some guidelines to erase all Kamailio files from the system?
> > I'm sure its a build problem.
>
>
> Sorry, this depends where you installed the server, you basically need to
> track all files that were installed. I missed the "make uninstall" also a
> few times. :/
>
>
> But you could just try to install the new version in another directory,
> like /tmp/kamailio or something like this. If you then configure the server
> to only use modules from this dir it should be not conflict with the old
> installation.
>
>
> Henning
>
I am running SER/rtpproxy on some systems with
multiple cores and multiple ethernet interfaces.
I am running multiple copies of rtpproxy, each with
an IP address associated with a particular interface,
as in one rtpproxy is using em1 which is 208.33.44.3
and em2 is 208.33.44.4 which another rtpproxy listens
at, while ser listens in em0 at 208.33.44.2. These
three (actually ten) interfaces pass through ethernet
switches and then on to a router with a default gateway
of 208.33.44.1 and then off to the Internet or wherever.
(In my setup, each rtpproxy has a second interface pointing
into a private network, so only one side of a call appears
on any given interface and five face toward the Internet.)
SER+rtpproxy correctly divides RTP traffic for calls to
208.33.44.3 and 208.33.44.4 in the SDP payloads, and
callers send their RTP packets to the two rtproxies
IP addresses without problems. That part works just fine.
The problem is that rtpproxy is using the default route
for transmission, so everything that goes towards the
Internet goes out via em0, so em1 and em2 don't transmit
anything. This means that this one link is overwhelmed
by the traffic that should have been divided across
multiple links.
Rtpproxy is allowing the local route table and the
default route to handle the outgoing packets and so
the packets are all going out the same interface.
In my case, em0 was 100Mbit/sec (completely sufficient
for SIP messages, while em1 and em2 were gigabit links
to handle RTP, but they are getting used only in one
direction. Even if em0 was upgraded to Gbit, it would
still not be able to handle the potential for 2+ Gbit
of data from the rtproxies.
So, is there any way to force rtproxy to perform what
I believe they used to call "reflector mode", where
the answers go out the same interface they come in on?
(This used to be common on UDP-based NFS file servers.)
A number of communication programs also allow this
behavior as an option, but rtpproxy doesn't appear
to do it or it isn't turned on by default.
Because the destinations for the RTP are out on
the Internet somewhere, I don't believe I can solve this
via "route" or "ifconfig" settings, but would be
delighted if someone knows different or knows of
a setting change in rtpproxy that will do this
so that load balancing will work.
The OS in question is FreeBSD 7.2, but I have the same
issue on 6.2 and 6.3.
Thanks in advance!
That said, this page does speak to a number of the concerns, even if
somewhat vaguely:
http://sip-router.org/answers/
>
>> I am not clear on what exactly the SIP-Router project is, nor how it
>> will affect me? Is Kamailio eventually being replaced with SIP-Router
>> once all the functionalities have been rolled in? Will my entire
>> configuration script need to be reworked to eventually operate with
>> this new platform? Can anyone explain to me why this merge is
>> happening so soon after Kamailio split with OpenSIPS? All this
>> turmoil feels very destructive to me.
>
> This is a confusing topic, and could definitely use additional discussion
> and official clarification.
>
> I have been made to understand that Kamailio will not be replaced by
> sip-router. It is rather that sip-router is a a "common core" that
> integrates both Kamailio and sip-router modules, and that both projects
> will share this common core going forward, but remain distinct. The
> analogy provided to me by Daniel was of Linux kernels vs. distributions.
>
> This seems fair enough, but it doesn't adequately explain who sip-router
> is intended for and who Kamailio is intended for, and who should use which
> and why and under what circumstances. It does also seem that most of the
> development and documentation energy is being concentrated into sip-router
> integration, which leaves open the question of what the future of Kamailio
> is.
>
> This definitely needs to be discussed more. Thank you very much for
> raising it, Geoffrey.
>
> --
> Alex Balashov
> Evariste Systems
> Web : http://www.evaristesys.com/
> Tel : (+1) (678) 954-0670
> Direct : (+1) (678) 954-0671
> Mobile : (+1) (678) 237-1775
>
>
>
> _______________________________________________
> Kamailio (OpenSER) - Users mailing list
> Users(a)lists.kamailio.org
> http://lists.kamailio.org/cgi-bin/mailman/listinfo/users
> http://lists.openser-project.org/cgi-bin/mailman/listinfo/users
>
--
Alex Balashov
Evariste Systems
Web : http://www.evaristesys.com/
Tel : (+1) (678) 954-0670
Direct : (+1) (678) 954-0671
Mobile : (+1) (678) 237-1775
Hey guys,
First off thanks for any help in advanced! I'm saving a nat branch flag
into usrloc when the user registers, now each user has two registrations to
the same username in usrloc, upon lookup("location"), inside of
branch_route[] it appears only one of them have the nat branch flag set,
while the other does not? Is there any case where a branch flag could
somehow get mangled, or what to look for, or what may cause this? Also one
other thing from a database outlook (the cflag) is equivalent on both AOR's.
However again at the time of branch_route, one flag is i.e. 000800, and the
other 000820, any thoughts are appreciated, thanks!
2009/3/23 Alexandr Dubovikov <shurik(a)start4.info>:
> On Mon, Mar 23, 2009 at 11:15:52AM +0100, Iñaki Baz Castillo wrote:
>> 2009/4/1 Alexandr Dubovikov <shurik(a)start4.info>:
>> > On Fri, Mar 20, 2009 at 10:21:20PM +0100, Andreas Heise wrote:
>> >> Hello Iñaki,
>
> Hi Inaki,
>
>> >
>> > Hi all,
>> >>
>> >> you should ask Alexandr he has introdused this feature with rev5452,
>> >> but I'm not sure if he is on the lists all the time, so I'll forward your
>> >> question
>> >> to him...
>> >
>> > sorry, I am currently offline and couldn't answer directly to the list.
>> >
>> > anyway. I agreed, 180 seconds it's too big ping interval, but it couldn't be
>> > also less 32 seconds.
>> >
>> >
>> > http://www.ietf.org/rfc/rfc3261.txt
>> >
>> > 17.2.2 Non-INVITE Server Transaction
>> >
>> >
>> > T1 = 500 ms.
>> >
>> >
>> > Timer J 64*T1 for UDP Section 17.2.2 Wait time for
>> > 0s for TCP/SCTP non-INVITE request
>> >
>> > so, the timer for "completed" is 32 seconds and couldn't be less.
>> >
>> >
>> > of course, you can use less value, but on own risk :-)
>> > RFC 3261 are not recommendet it :)
>>
>> Thanks for pointing it out.
>> However, if a gw takes so long time (~32 seconds) to respond I prefer
>> to consider it as "offline" :)
>
> don't forget to disable ping for this gateway, because finaly you will have
> a DDOS effect of the "offline" gw. :-)
Well, If the host is already offline then it doesn't matter to kill it more XD
Thanks.
--
Iñaki Baz Castillo
<ibc(a)aliax.net>
Hi,
I am a new user of openser. I have installed and configured openser on my
local system. when i am trying to add a new user in database i am getting
following error:
[root@Ajay-PC sbin]# openserctl add ajay ajay ajay@localhost
database engine 'MYSQL' loaded
Control engine 'FIFO' loaded
is_user: user counter=0
check_db_alias: alias counter=0
ERROR 1054 (42S22) at line 1: Unknown column 'phplib_id' in 'field list'
ERROR: introducing the new user 'ajay' to the database failed
Kindly help me.
Thanking you.
--
View this message in context: http://www.nabble.com/Problem-in-adding-new-user-in-database-tp24307793p243…
Sent from the OpenSER Users Mailing List mailing list archive at Nabble.com.