Hello guys, I have been trying to get openser implemented with SEMS the past
day or so although I keep running into problems, any advice on setting up
openser with SEMS? (Sip Express Media Server).
Although here is a syntax error straight from the documentation:
...
modparam("tm","tw_append","append1:Email=avp[i:12];UA=hdr[User-Agent]")
modparam("tm","tw_append","append2:body=msg[body]")
...
t_write_fifo("voicemail/append1","/tmp/appx_fifo");
...
t_write_unix("logger/append2","/var/run/logger.sock");
Is the syntax for "t_write_unix". Although in my code I have the following:
modparam("tm", "tw_append", "vm_email:P-Email-Address=avp[i:34]")
Just like the example..... although I get error:
0(57475) ERROR:tm:parse_tw_append: parse error in
<vm_email:P-Email-Address=avp[i:34]> around position 25(a)
Any idea on what is going on? Thanks, all help is appreciated!
--
Brandon Armstead
Hello, list!
I currently use the debian package of OpenSER (v1.1.0-5), and I see a behaviour I do not explain with RFC's and OpenSER Documentation.
I have registered several User-Agent under the same username/domain couple in an OpenSER instance with the registrar module. When I try to call "username@domain", I expected OpenSER to call the contact with the highest q-value, and keep the other contacts into transaction branches for later use (as said in the usrloc module's documentation, in the "lookup" function description).
In spite of this, I observed that OpenSER called each contact in a parallel way, and the first UA that responds a 200 is the one that receive the final ACK, other branches are closed.
Is it wanted / intentional / normal ? Is the documentation up to date?
Whatever will your answer be, thank's a lot!
Jean-François Smigielski.
________________________________________________________________________
iBELGIQUE, exprimez-vous !
http://web.ibelgique.com/
Hello, I get syntax error on the following two commands, does anyone have
any idea, thanks:
avp_db_load("$ru", "i:34/$email_scheme")
modparam("tm", "tw_append", "vm_email: P-Email-Address=avp[i:34]")
--
Brandon Armstead
I am currently running load testing on SER and have a question
regarding multiple contacts registered with the same IP and port
number but different contact name. Will this work? Does each contact
ID have to be defined with a different port number to establish a
session?
For example:
Contact : sip:loadtest0100@10.180.0.71:10140
Contact : sip:loadtest0101@10.180.0.71:10140
Contact : sip:loadtest0102@10.180.0.71:10140
Contact : sip:loadtest0103@10.180.0.71:10140
Hi,
Anyone know how openser works with following:
1. Forking. Often times, I may have one account register with multiple
SIP client, does it support that? I don't seem to make it working.
2. Unregister and register. Nowadays, there are a lot of SIP client
will send an Unregister message to the SIP server and after received
response, it will send the new registration again, this is because when a
client behind the same router reboot multiple times, it might have
registration in the SIP server and only the latest one reflect the current
registration and all previous registration may not be valid as the router
may already close the mapping port, so in RFC3261, it defines client can
send a registration with contact head contains "*" and expeiration timer is
0 to unregister the previous registration. It does not seem that OPENSER
can handle it. I know SER handles it.
3. Can I specify OPENSER to only send the INVITE (or other SIP request)
to the latest AOR for a particular account?
Thanks in advance.
Bill
I installed the openser.cfg at
http://voip-info.org/wiki/view/OpenSER+And+Mediaproxy (modified to
authenticate against my domain, and to use exit instead of break). Calls
between two phones on the domain work, but there is a lot of jitter. Well,
not exacty jitter, but kind of a pulsing sound in the background of the
call. Also, will occasionally drop out, and then come back after a little
bit.
Hi everyone.
I'm facing this problem when configuring a Ser as a proxy and a registrar.
I have the Ser and a IP pbx: the pbx registers itself on the proxy as
pbx@<pbx_ip> (Contact header).
I had to manage INVITEs messages for incoming call that are going
towards the Ser, sent by UACs in the network.
INVITEs are coming in the form <telephone_number>@<ser_ip> (Req-URI):
Ser had to decide if <telephone_number> is managed by the IP pbx (it has this information) and eventually, do a rewrite uri to
<telephone_number>@<pbx_ip> and then forward the message to pbx.
So far, it's all ok.
Unfortunately, i have to handle also this situation: the ip pbx is down.
At the present, the INVITE is forwarded towards the pbx, the ip pbx does not respond and after 10 seconds the requesting UACs sends a CANCEL.
This latency is not acceptable.
I'm looking for a way to implement this:
after the rewrite uri to <telephone_number>@<pbx_ip>, look if I have a
contact in usrloc with <pbx_ip> (i.e. the pbx is still registered): if
not, send immediately a response 4xx to UAC.
The problem is that i'm not able to access the usrloc db from the
routing script... any hint?
Thanks a lot...
I have openser configured with asterisk as following:
If a sip REGISTER packet arrives,
fix_nated_register()
save("location")
exit;
If a sip packet arrives at openser with a source ip != asterisk ip,
t_relay(to asterisk)
If a sip packet arrives from asterisk,
lookup("location");
force_rport()
fix_nated_contact()
t_relay();
The asterisk dial plan says the following:
if inbound traffic arrives from openser's ip address, Dial(
SIP/dstuser@dstdomain).
phones a and b are on separate networks behind a firewall, openser and
asterisk are on public ip addresses.
The sip traffc seems to work just fine. I.e. all the handshakes seem to be
happening as they should.
However, rtp traffic does not.
Whether audio traffic will transmit or not is a crap shoot.
Sometimes, if a calls b, b hears a but not vice versa, while if b calls a,
audio is two way.
In this case, one audio stream is going through asterisk, the other is being
directed to go point to point.
Sometimes, a calls b and b hears a, and a hears b for a second but a second
INVITE comes to phone B that causes it to redirect rtp to be point to point.
Sometimes there is no audio.
Sometimes, everything works fine.
At one point, rtp from a was going to asterisk, but asterisk was not sending
the rtp on to b, and b was trying to send traffic point to point.
Dear Witt,
Thank you very much for your help before.
Witt, may I ask you again,please?
I have downloaded ssldump.tar.gz packet and installed it.
As the INSTALL note in ssldump, I have installed libcap.tar.gz (from www.tcpdump.org) and openssl.
I have installed it by using commands:
# tar -zxvf ssldump-0.9b3.tar.gz
# cd ssldump-0.9b3
# ./configure --with-openssl
#make
#make install
All installation process can run sucessfully (nothing error message shown).
But, why when I try to run ssldump, I got nothing. There is any mesage shown when I tried to run
#ssldump -i eth0
What happens? Did I do wrong installation process? or I have missed something?
I do hope you can help me.Please help me..
Thank you very much for your help
Regards,
Ferianto
Note:
Actually, when I run ./configure --with-pcap, I got error message. The error message said that there is no pcap file. Why?
Is lipcap packet that I got from www.tcpdump.org (as I read from ssldump INSTALL), same with pcap file ? or I have to download pcap ?
Steffen Witt <witt.steffen(a)googlemail.com> wrote:
Hello Ferianto,
TLS is used for encrypting your SIP signaling, not the media data.
You should install "ssldump" on your server to see each step of the
TLS handshake.
Maybe you have a closer look to this short SSL tutorial:
http://www.eventhelix.com/RealtimeMantra/Networking/SSL.pdf
Best regards
Steffen
2006/10/30, Ferianto siregar :
> Dear all,
>
> Thank you for this chance.
> All, I have tried to learn about how can TLS can secure communication in
> VoIP.
> I must to understand it in order to responsible my openser server (working)
> with TLS to my lecturer.
> But, I have still doubt.
> Actually, what is encrypted in TLS system so the communication become
> secure? Does it encrypt the data (voise packet)? or The traffic (so the
> sniffer can not get the data? OR anything else?
>
> I do hope anybody can help me :)
> Thanks,
>
>
> Ferianto
>
>
> Note: I cryptography tutorials, I read that the cryptography system secure
> the message (text) by encrypting it using public key and private key. So,
> How about TLS in VoIP?
>
>
> ________________________________
> Everyone is raving about the all-new Yahoo! Mail.
>
>
> _______________________________________________
> Users mailing list
> Users(a)openser.org
> http://openser.org/cgi-bin/mailman/listinfo/users
>
>
>
---------------------------------
Get your email and see which of your friends are online - Right on the new Yahoo.com
Hi,
I changed my linux distro to debian (from slack) so i can install openser from packages, but aptitude failed to install avp_radius. When I install from source, avp_radius.so is created, after tweaking the Makefile. What is the problem, and how do i get avp_radius module when i install from aptitude/apt-get/dpkg ?
best regards, Zoran