Hi all,
can anybody tell where we have to put the routing code in
openser.cfg for CRBT(caller ring backtone ) application.
A------------------------------------------make call -------------->B
same time
A------------------------------------------------------------------------->C
in "C" we are playing prompts.
Regards
Amit
Hi ,
As a part of my MS project I proposed a 3-party authentication protocol which can be used in SIP too. for SIP. The proxy server utilizes this protocol to authenticate the UC's Invite,using the Registrar server. Now, I have to estimate the totoal overhead of the scheme on a normal SIP call. I searched dozens of SIP, UMTS, GSM related papers to find a simulator or emulator which could be used to to achive the goal but I couldn't get any solutions as I need to consider execution of the RSA,AES and SHA on the agent's processor, I also need the solution to answer me with not too much programming due to my time constraints. I browsed the Resirocate project docs but I don't think then can answer me in time. I am also not sure whether SIPp, MS LIve communication server and SERcan help me? Could you please suggest me any rapid solutions? Any simulation techniques?any piece of software or programmining interface? By the way I am a little familiar with opnet as I did a tiny
project with it. Doest it help me?
Thanks in advance.
Samira.
---------------------------------
Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello,
I'm quite unsure about openser's hangling of message processing. So can
anybody answere following 2 questions please ?
1. When multiple messages come in, lets say 3 INVITEs. Will openser fork
one child for each message, run the route script and terminate the child
after that? Or will openser throw all 3 INVITEs at the same time into
the route script?
2. If the 3 INVITEs share the same route script, are they handled
sequentially? What happens to the AVPs and $var variables? Are they
gonna mixed?
regards
Helmut
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFHVRWA4tZeNddg3dwRAkQZAJ9Ruy3kr2RzODDlm55uZ4Y6yZTcPACgrl+v
qmOW524gE6vgsWNgOalpfW4=
=oR4D
-----END PGP SIGNATURE-----
Hi all,
what is the different between napting interval done by nathelper-module and
the other ones done by mediaproxy-module ?
Can I set both of them at the same time ?
regards,
daniel
--
Daniel Grotti
________________________
e-mail : d.grotti(a)gmail.com
Good day friends,
I was trying to install Openser and Asterisk on same server. I need some configuration or information if anyone can point me.
I need to forward only International call to the Asterisk but the same time if registered accounts of openser want to make call between them the can call each other without routing to asterisk.
Could you please give us the right parameters of openser.cfg to route calls between openser accounts?
Regards,
www.Go4Calls.Com
VoIP Forums
_________________________________________________________________
Discover the new Windows Vista
http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE
Hi, All
I am with the OPENSER running some time without problems.
Yesterday, he went not to complete any call, and the system had the following message log:
ERROR:registry:update_contacts: invalid cseq for aor
ERROR:registry:update_contacts: invalid cseq for aor
ERROR:registry:update_contacts: invalid cseq for aor
for all numbers
Thanks
Hi,
I need to configure my ser / openser in such a way that these sites should
be able to call each other :
Site A -> 01113, 01114
Site B -> 02113, 02114
Site C -> 03113, 03114
User in Site A (Dial 02113 or 03113) --------> SER / OpenSER --------> User
in Site B or Site C
and they should be able to call within site :
User in Site A (Call 114) ----------> SER / OpenSER ---------> User in
Site A
here 01, 02, 03 are my site code so site code is only needed when users want
to call between sites. If they are calling within same site no need to dial
site id.
please advise and give some links.
Thanks,
arun
Hi All,
This is specifically for the SER/OpenSER developers, but I'm not a serdev
list member so I'm posting here.
I've been using SER since version 0.8.X and I'm still running
0.8.14production for my company PBX to this day.
I was very excited as version 2 became a release candidate and I downloaded
it for testing. I was pretty disappointed with one aspect of the new data
model and I'm requesting that the developers consider a further revision on
the data model.
Basically, taking all fields out of the subscriber table like Last_name,
first_name, email, timezone, rpid/asserted identity, etc, etc is not the
greatest idea. It's a better idea from a database architecture and
performance perspective to keep adding columns into that table for data that
has a 1 to 1 relationship with a user, and that is common in > 90% of SER's
use cases (i.e. corporate, carrier/VSP.) I would suggest adding
voicemail_password, and maybe every other field that is being added into the
default attributes script that I saw in CVS recently. If you already know
what attributes a user has (and they have a 1 to 1 relationship), then its
far better from a db performance perspective to keep these attributes in the
user table. I know that the code becomes more complicated, but I think it
may be a tradeoff worth discussing.
See this discussion (
http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:1067808…)
between Oracle users and Tom, (an Oracle engineer/architect.) The
full
text of this discussion is very informative and I highly recommend people
read it through.
Tom's conclusion is that the type of data model being discussed, and now
being used in SER fails for all but the most trivial of applications. Maybe
SER *by itself* qualifies as "trivial" from a database architect's
perspective, but think about things like Asterisk integration, which is
quite common. You quickly run into some very nasty queries . . .
Please note that I am not a software developer nor a database engineer, just
a user who reads a lot, so I'm open to being the ignorant one here, but I
thought that this should be discussed among users and developers.
Thanks for considering,
Mahatma
http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:1067808…
The following is an excerpt from the above link:
Here is a excerpt from my forthcoming book where I talk about this
(and show you how
ugly, hard and inefficient queries against your very flexible model will be)
(2)Do not use Generic Data Models
Frequently I see applications built on a generic data model for
"maximum flexibility" or
applications built in ways that prohibit performance. Many times -
these are one in the
same thing! For example, it is well known you can represent any object
in a database
using just four tables:
Create table objects ( oid int primary key, name varchar2(255) );
Create table attributes
( attrId int primary key, attrName varchar2(255),
datatype varchar2(25) );
Create table object_Attributes
( oid int, attrId int, value varchar2(4000),
primary key(oid,attrId) );
Create table Links ( oid1 int, oid2 int,
primary key (oid1, oid2) );
That's it - no more CREATE TABLE for me! I can fill the attributes
table up with rows
like this:
insert into attributes values ( 1, 'DATE_OF_BIRTH', 'DATE' );
insert into attributes values ( 2, 'FIRST_NAME', 'STRING' );
insert into attributes values ( 3, 'LAST_NAME', 'STRING' );
commit;
And now I'm ready to create a PERSON record:
insert into objects values ( 1, 'PERSON' );
insert into object_Attributes values( 1, 1, '15-mar-1965' );
insert into object_Attributes values( 1, 2, 'Thomas' );
insert into object_Attributes values( 1, 3, 'Kyte' );
commit;
insert into objects values ( 2, 'PERSON' );
insert into object_Attributes values( 2, 1, '21-oct-1968' );
insert into object_Attributes values( 2, 2, 'John' );
insert into object_Attributes values( 2, 3, 'Smith' );
commit;
And since I'm good at SQL, I can even query this record up to get the
FIRST_NAME and
LAST_NAME of all PERSON records:
ops$tkyte@ORA920> select
max( decode(attrName, 'FIRST_NAME', value, null )) first_name,
2 max( decode( attrName, 'LAST_NAME', value, null ) ) last_name
3 from objects, object_attributes, attributes
4 where attributes.attrName in ( 'FIRST_NAME', 'LAST_NAME' )
5 and object_attributes.attrId = attributes.attrId
6 and object_attributes.oid = objects.oid
7 and objects.name = 'PERSON'
8 group by objects.oid
9 /
FIRST_NAME LAST_NAME
-------------------- --------------------
Thomas Kyte
John Smith
Looks great, right? I mean, the developers don't have to create tables
anymore, we can
add columns at the drop of a hat (just requires an insert into the
ATTRIBUTES table). The
developers can do whatever they want and the DBA can't stop them. This
is ultimate
"flexibility". I've seen people try to build entire systems on this model.
But, how does it perform? Miserably, terribly, horribly. A simple
"select first_name,
last_name from person" query is transformed into a 3-table join with
aggregates and all.
Further, if the attributes are "NULLABLE" - that is, there might not
be a row in
OBJECT_ATTRIBUTES for some attributes, you may have to outer join
instead of just joining
which in some cases can remove more optimal query plans from consideration.
Writing queries might look pretty straightforward, but it's impossible
to do in a
performant fashion. For example, if we wanted to get everyone that was
born in MARCH or
has a LAST_NAME = 'SMITH', we could simply take the query from above
and just wrap an
inline view around that:
ops$tkyte@ORA920> select *
2 from (
3 select
max(decode(attrName, 'FIRST_NAME', value, null)) first_name,
4 max(decode(attrName, 'LAST_NAME', value, null)) last_name,
5 max(decode(attrName, 'DATE_OF_BIRTH', value, null))
date_of_birth
6 from objects, object_attributes, attributes
7 where attributes.attrName in ( 'FIRST_NAME',
'LAST_NAME', 'DATE_OF_BIRTH' )
8 and object_attributes.attrId = attributes.attrId
9 and object_attributes.oid = objects.oid
10 and objects.name = 'PERSON'
11 group by objects.oid
12 )
13 where last_name = 'Smith'
14 or date_of_birth like '%-mar-%'
15 /
FIRST_NAME LAST_NAME DATE_OF_BIRTH
-------------------- -------------------- --------------------
Thomas Kyte 15-mar-1965
John Smith 21-oct-1968
So, it looks "easy" to query, but think about the performance! If you
had a couple
thousand OBJECT records, and a couple tens of thousands of
OBJECT_ATTRIBUTES - Oracle
would have to process the entire inner group by query first and then
apply the WHERE
clause.
This is not a made up data model, one that I crafted just to make a
point. This is an
actual data model that I've seen people try to use. Their goal is
ultimate flexibility.
They don't know what OBJECTS they need, they don't know what
ATTRIBUTES they will have.
Well - that is what the database was written for in the first place:
Oracle implemented
this thing called SQL to define OBJECTS and ATTRIBUTES and lets you
use SQL to query
them. You are trying to put a generic layer on top of a generic layer
- and it fails each
and every time except for the most trivial of applications.
Hi Douglas,
You will need to upgrade to 1.2.2.
This issue was fixed starting with 1.2.0.
Regards,
Ovidiu Sas
On Dec 3, 2007 3:12 PM, Douglas Garstang <dougmig33(a)yahoo.com> wrote:
>
> Ok, I can accept that it might be a timeout since it doesn't happen every
> single time.
> 1/5s seems awefully short though. How can I increase this? I can't find
> anything in the docs about increasing this timeout...
>
> Doug.
>
>
>
> ----- Original Message ----
> From: Ovidiu Sas <osas(a)voipembedded.com>
> To: Tavis P. <tavis.lists(a)galaxytelecom.net>
> Cc: users(a)openser.org
> Sent: Monday, December 3, 2007 12:01:53 PM
> Subject: Re: [OpenSER-Users] OpenSER Sending INVITE Twice!?
>
> Right, 1/5 is not 1/2 :)
>
> -ovi
>
> On Dec 3, 2007 10:10 AM, Tavis P. <tavis.lists(a)galaxytelecom.net> wrote:
> >
> > It may also be related to the timer accuracy of OpenSER 1.1
> >
> > If the T1 interval is 500ms but OpenSER 1.1 doesn't support sub-second
> > timer accuracy it is probable to see it retransmitting in situations where
> > T1 hasn't been reached yet
> > It shouldn't cause any problems though, just a few erroneous retransmits
> > from time to time
> >
> > tavis
> >
> > Ovidiu Sas wrote:
> > Hi Douglas,
> >
> > You posted this questionawhileago and by checkingthe archives
> > Irealised that my answer was truncated somehow ...
> > Here's the full answer:
> >
> > This is the way that is supposed to work. If nothing is received from
> > upstream, the request is resent.
> > >From the rfc3261:
> >
> > 17.1.1.1 Overview of INVITE Transaction
> >
> > The INVITE transaction consists of a three-way handshake. The client
> > transaction sends an INVITE, the server transaction sends responses,
> > and the client transaction sends an ACK. For unreliable transports
> > (such as UDP), the client transaction retransmits requests at an
> > interval that starts at T1 seconds and doubles after every
> > retransmission. T1 is an estimate of the round-trip time (RTT), and
> > it defaults to 500 ms.
> >
> >
> > It seems that your upstream UAS is slowin sending out the 100 Trying
> > and therefor openSER is resending out the INVITE.
> >
> >
> > Regards,
> > Ovidiu Sas
> >
> > On Dec 3, 2007 12:35 PM, Douglas Garstang <dougmig33(a)yahoo.com> wrote:
> >
> >
> > I have an OpenSER 1.1 install here, and for some reason, OpenSER is
> sending
> > the INVITE message twice to the upstream host.
> >
> > Asterisk OpenSER Provider
> > -----------------------------------------------
> > | ---INVITE---> | |
> > | | ---INVITE---> |
> > | | ---INVITE---> |
> > | | <---TRYING--- |
> > | | <---TRYING--- |
> >
> > The time difference between them is about 1/5 of a second. Nothing is
> > received between the first and second INVITE's.
> >
> > Why is OpenSER doing this?
> >
> > I have xlog() statements everywhere in openser.cfg, and OpenSER is only
> > logging ONE outgoing INVITE message, eventhough it's sending two. It is
> > logging the multiple TRYING messages that come back however.
> >
> > OpenSER is not calling failure_route because nothing is logged in there.
> > What could be going wrong? Why is it doing this?
> >
> > Doug.
> >
> > ________________________________
> > Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it
> > now.
> > _______________________________________________
> > Users mailing list
> > Users(a)lists.openser.org
> > http://lists.openser.org/cgi-bin/mailman/listinfo/users
> >
> >
> >
> > _______________________________________________
> > Users mailing list
> > Users(a)lists.openser.org
> > http://lists.openser.org/cgi-bin/mailman/listinfo/users
> >
> >
> >
> >
> > _______________________________________________
> > Users mailing list
> > Users(a)lists.openser.org
> > http://lists.openser.org/cgi-bin/mailman/listinfo/users
> >
> >
>
> _______________________________________________
> Users mailing list
> Users(a)lists.openser.org
> http://lists.openser.org/cgi-bin/mailman/listinfo/users
>
>
> ________________________________
> Get easy, one-click access to your favorites. Make Yahoo! your homepage.
I have an OpenSER 1.1 install here, and for some reason, OpenSER is sending the INVITE message twice to the upstream host.
Asterisk OpenSER Provider
-----------------------------------------------
| ---INVITE---> | |
| | ---INVITE---> |
| | ---INVITE---> |
| | <---TRYING--- |
| | <---TRYING--- |
The time difference between them is about 1/5 of a second. Nothing is received between the first and second INVITE's.
Why is OpenSER doing this?
I have xlog() statements everywhere in openser.cfg, and OpenSER is only logging ONE outgoing INVITE message, eventhough it's sending two. It is logging the multiple TRYING messages that come back however.
OpenSER is not calling failure_route because nothing is logged in there.
What could be going wrong? Why is it doing this?
Doug.
____________________________________________________________________________________
Looking for last minute shopping deals?
Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping
