Hi, all!
Preamble:
I got new equipment - videophone. I tried to make it work with OpenSER
(version 1.1.0).
Schema:
VIDEOPHONE -> OpenSER -> VIDEOPHONE
I tried to make simple call.
It works but not behind NAT.
With rtpproxy OpenSER logs show such messages:
ERROR: force_rtp_proxy2: can't extract media port from the message
Calls last only few seconds than it get disconnected.
With mediaproxy OpenSER logs show following:
Oct 16 18:01:22 sip mediaproxy[20251]: error: uncaptured python
exception, closing channel <rtphandler.CommandHandler connected '' at 0x68
2ecc> (exceptions.ValueError:invalid literal for int(): 5012/1
[/usr/lib/python2.3/asyncore.pyc|readwrite|86] [/usr/lib/python2.3/asyncore
.pyc|handle_read_event|390]
[/usr/local/mediaproxy/modules/rtphandler.py|handle_read|441]
[/usr/local/mediaproxy/modules/rtphandler.py|pro
cess|531] [/usr/local/mediaproxy/modules/rtphandler.py|__init__|988]
[/usr/local/mediaproxy/modules/rtphandler.py|setCaller|880])
I found that source of these troubles is that videophone generates
INVITE`s in different format than for example Sipura does.
Media streams are described like these:
m=audio 5012/1 RTP/AVP 98 0 8.
m=video 5010/1 RTP/SAVPF 99.
As you can see, sub-field "port" is not in usual format, but it contains
number of ports according to RFC2327:
m=<media> <port>/<number of ports> <transport> <fmt list>
So, I spended few hours examinating sources of rtpproxy and mediaproxy
modules and made small trick (patch) to mediaproxy module.
I didn't understand logic of rtpproxy module because lack of time.
Here it is:
--- ./modules/mediaproxy/mediaproxy.c 2006-03-24 21:04:56 +0300
+++ ./modules/mediaproxy/mediaproxy_new.c 2006-10-17 17:53:54 +0400
@@ -825,7 +825,7 @@
getMediaStreams(str *sdp, str *sessionIP, StreamInfo *streams, int limit)
{
str tokens[2], block, zone;
- char *ptr, *sdpEnd;
+ char *ptr, *sdpEnd, *tmpptr;
int i, count, streamCount, result;
sdpEnd = sdp->s + sdp->len;
@@ -845,6 +845,9 @@
"line in SDP body\n");
return -1;
}
+
+ tmpptr=strchr(tokens[1].s,'/');
+ if ((tmpptr-tokens[1].s)<tokens[1].len )
tokens[1].len=tmpptr-tokens[1].s;
streams[i].type = tokens[0];
streams[i].port = tokens[1];
It works for me, but I know it is not correct solution.
I hope my small investigation would be useful.
Thanks for your attention.
With best regards,
Artur Yakupov
Dear all,
Thank you very much for this chance. Thank you very much for time.
All, I have a problem in understanding the output when running openserctl add command (for adding new user).
After, I add new user (success), Why this message always appear?
What does it mean?
"0\n".
"1.00\n".
"0\n".
"128\n".
"4294967295\n";
Because I see that this number never changed. I always get this number when success to add new user.
I do hope anybody can give me information, what do those numbers mean?
Thanks
Regards,
FERIANTO
---------------------------------
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail.
Hello Iptel users.
I'm having a problem with my mediaproxy box. I have a NAT'd client (a softphone) calling to a SIP-H323 conversor through my SER-Mediaproxy box and then to a Cisco Call Manager (it's a very weird configuration). Anyway, the problem presents when the 200 - OK message arrives from my conversor, the SDP part indicate the media address (10.10.148.230) and port (16952) where the remote endpoint will listen RTP, despite of this the mediaproxy open a session for the port 17360. Please check the log from the mediaproxy and the execution of the "sessions.py" utility, plus the OK message.
U 100.100.154.36:5060 -> 100.100.148.246:5060
SIP/2.0 200 OK.
Via: SIP/2.0/UDP 100.100.148.246;branch=z9hG4bK67ea.aa12dc97.0.
Via: SIP/2.0/UDP 192.168.1.101;received=100.100.148.248;rport=46096;branch=z9hG4bKc0a801650000001d4533a0ee00007c2900000011.
Record-Route: <sip:100.100.154.36;ftag=2214787970;lr=on>.
Record-Route: <sip:7072408196@100.100.148.246:5060;nat=yes;ftag=2214787970;lr=on>.
From: "unknown"<sip:5502408196@sip.desa.mydomain.net>;tag=2214787970.
To: <sip:7072408196@sip.desa.mydomain.net>;tag=2260729709.
Call-ID: A2304B7B-AA05-42C5-A467-C16B1BD9102A(a)192.168.1.101.
Cseq: 2 INVITE.
Date: Mon, 16 Oct 2006 15:10:37 GMT.
Server: siph323.
Content-Type: application/sdp.
Content-Length: 162.
Contact: sip:7072408196@100.100.154.119.
.
v=0.
o=100.100.154.119 1 505533324 IN IP4 100.100.154.119.
s=SIP Library call.
c=IN IP4 100.100.148.230.
t=3370000237 0.
m=audio 16952 RTP/AVP 0.
a=rtpmap:0 PCMU/8000.
Oct 16 12:12:00 sip mediaproxy[18289]: lookup A2304B7B-AA05-42C5-A467-C16B1BD9102A(a)192.168.1.101 100.100.148.230:16952:audio 100.100.154.36 sip.desa.mydomain.net local sip.desa.mydomain.net unknown siph323 info=from:5502408196@sip.desa.mydomain.net,to:7072408196@sip.desa.mydomain.net,fromtag:2214787970,totag:2260729709
Oct 16 12:12:00 sip mediaproxy[18289]: execution time: 0.48 ms
Oct 16 12:12:00 sip mediaproxy[18289]: session A2304B7B-AA05-42C5-A467-C16B1BD9102A(a)192.168.1.101: caller signed in from 100.100.148.248:49154 (RTP) (will return to 100.100.148.248:49154)
Oct 16 12:12:00 sip mediaproxy[18289]: session A2304B7B-AA05-42C5-A467-C16B1BD9102A(a)192.168.1.101: called signed in from 100.100.148.230:17360 (RTP) (will return to 100.100.148.230:17360)
[root@sip ser]# /usr/src/mediaproxy-1.7.2/mediaproxy/sessions.py
Caller Via Called Status Duration Codec Type Traffic
------------------------------------------------------------------------------------------------------------------------
100.100.148.248:49154 - 100.100.148.246:60022 - 100.100.148.230:17360 active 0'39" G711u Audio 336.91k/669.53k/332.62k
Total traffic: 79.69kbps/78.12kbps/157.81kbps (in1/in2/out)
Session count: 1
Proxy version: 1.7.2
Doing more debugs i discover the RTP traffic coming from the 100.100.148.230 it has source port 17360, but the listening port is not the same (i guess is the one indicated in the SDP). Why mediaproxy assumes the 17360 port is for listening too and ommit the SDP 16952 port ?
Is this maybe an asymmetric regarding the RTP media?. If so, i added this line into the rtp-asymmetric-clients file:
#
# Lines starting with a '#' character or empty lines are ignored.
# Put a User-Agent name or regular expression per line.
# Check is case insensitive.
#
# This file should only list SIP clients that are asymmetric regarding
# the RTP media streams. For clients that are asymmetric regarding the SIP
# signaling, use the equivalent SIP version of this file.
# Clients that are asymmetric for both SIP signaling as well as RTP media
# streams should go in both files.
#
siph323
But not solve the problem.
Can someone help me here?
Thanks in advance
Regards
Ricardo Martinez.-
RedVoiss.
Hello List,
Can any one please suggest , how to use AAA in ser with the help of free
radius. I had included the radius module in the ser.cfg file, I would like
to know what other files needed in the configuration.
Is there any How to available to do that
Thanks
Lokesh
--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.408 / Virus Database: 268.13.4/480 - Release Date: 10/17/2006
Hi !
Which software I can use for service of I pre-paid
with openser?
Please help me !
Thanks.
Moacir.
_______________________________________________________
O Yahoo! está de cara nova. Venha conferir!
http://br.yahoo.com
On Wednesday 18 October 2006 05:03, senthil vinayagam wrote:
> Hi Richard,
>
> If you are using openser with TCP/TLS enabled then Openser will close the
> tcp/tls connection after couple of minutes of inactivity by the phone.
> Either we have to change the code of openser to increase the tcp
> inactivity timer or change phone to re-register every 80 seconds or so.
> Otherwise the tcp/tls connection shall get closed and no further calls can
> be placed after that.
>
> My suggestion would be just place a single call & observe. See if it
> happens of a single call.
No, it is not TLS enabled, so probably not that.
Also it is random, redial and it is fine... Happens about once every 10 or 20
calls.
Thanks
Richard
>
On 17/Oct/2006 18:08 Cesc wrote ..
> nice ... i will take a look.
> BTW, any experience with phones?
>
> Cesc
Hello,
I have used x-lite, wengo-ng and presence server and it works well even though there are some presence bugs in wengo-ng...
Best Regards,
Henri Keski-Sikkila
Hi,
Just wondering if someone is facing similar problem.
My setup is SER/MediaProxy/MYSQL on public IP, I have couple of dozen
SPA-1001 all over the internet.
The problem I'm facing is that many of these ATA SPA-1001 some times do not
ring, so try to ping (serctl ping sip:550505050@myserver.com) and i get
error 408.
When i restart these ATA everthing starts working but couple of hours later,
it's the same problem.
Not to mention that (serctl showdb) shows that sip client is registered,
however I still get 408.
I tried the following: (but no success)
1. SPA-1001 change Register Expires: from 3600 to 600
2. re-INVITE TO 30
Can someone help, or maybe there are some manufacturing bug on some linksys
SPA-1001, by the way all my ATA use uniform configs and firmware.
If someone is willing to look into the ngrap, i can show it to you.
Thanks,
Ali...
On 17/Oct/2006 18:08 Cesc wrote ..
> nice ... i will take a look.
> BTW, any experience with phones?
>
> Cesc
Hello,
I have used x-lite, wengo-ng and presence server and it works well even though there are some presence bugs in wengo-ng...
Best Regards,
Henri Keski-Sikkila
