Hello,
we are using an RTP Proxy from rtpproxy.org as media relay to establish
communication between our mobile phones. Of course, we are using the
kamailio rtpproxy module to modify the SDP payload and control the proxy.
In our Kamailio configuration, we have 1 kamailio configured as Proxy and
one kamailio configured as Registrar. So calls go through the Proxy and then
to the Registrar who will update the SDP header and select an available rtp
proxy.
We have noticed that sometimes, the rtp udp flow between the phones isn't
routed properly by the rtpproxies, ending in the communication drop (all the
SIP nego is working well, and the SDP are correctly patched with the rtp
proxy address and port).
Analyzing the RTP proxy packets, we have found that the Kamailio registrar
gives the Kamailio proxy ip address in the RTP proxy create session command,
but keeps the original sdp port.
command looks like this:
Uc96,101 DC -PbO~Wnm <proxy_ip> <port from original sdp phone packet>
PZU5OITCW;1
We are using rtpproxy_manage() without any flags.
It seems to us that this ip and port are used as default forward route as
long as the callee hasnt connected to the rtpproxy. *Is it correct?
*
If its true, Its seems to us that this cant work as we are mixing the proxy
sip address with a udp port open on the phone ? *Is our analysis correct ?*
Can we use some option in rtpproxy_manage to replace the proxy ip by the
phone ip as seen in the via route ?
Thx for your help
Giovanni
--
View this message in context: http://sip-router.1086192.n5.nabble.com/rtpproxy-module-question-tp148850.h…
Sent from the Users mailing list archive at Nabble.com.
Hi guys!
I am trying to configure kamailio with WSS.
We have trusted certificate installed SIP over TCP/TLS works fine.
But when I try WSS I got error:
ERROR: tls [tls_util.h:42]: tls_err_ret(): TLS read:error:14094419:SSL
routines:SSL3_READ_BYTES:tlsv1 alert access denied
ERROR: <core> [tcp_read.c:1303]: tcp_read_req(): ERROR: tcp_read_req: error
reading
Before above error it was 'bad certificate', so I have imported CA in the
firefox and now I get these errors..
I have tried sipml5 and tryit.jssip.net same issue with both clients, also
it seems I have these errors only when I use firefox, when I use chrome it
even doesn't show me an error..
Any ideas?
Thanks!
--
Regards,
Arsen.
Hello,
Sorry if this is a tired, worn question, but I've not dealt much with
Kamailio's SRV support before:
If a registrant has a contact binding whose domain component is subject
to an SRV lookup with load-balanced or weighted entries, how does one
solve the problem of ensuring that subsequent in-dialog requests go to
the same host as the initial INVITE? Does Kamailio offer some facility
for doing this? Is it somehow accommodated by SIP?
-- Alex
--
Alex Balashov | Principal | Evariste Systems LLC
1447 Peachtree Street NE, Suite 700
Atlanta, GA 30309
United States
Tel: +1-800-250-5920 (toll-free) / +1-678-954-0671 (direct)
Web: http://www.evaristesys.com/, http://www.csrpswitch.com/
Hi all,
Im struggling with incoming calls to webrtc clients.
I am trying to change INVITE's SDP using rtpengine from RTP/AVP or
RTP/SAVP to RTP/SAVPF towards my webrtc clients. This works well.
But if I send INVITE with SDP that includes RTP/AVP and RTP/SAVP as
well, rtpengine creates incorrect SDP. It includes twice m=audio
RTP/SAVPF and my WebRTC client does not like this INVITE and forbiddeds
it by 488 Not Acceptable Here.
Example of my incorrect INVITE :
Content-Type: application/sdp
Content-Disposition: session
Content-Length: 1393
v=0
o=SBC 1464238673 1464238674 IN IP4 <MyPublicIP>
s=SBC
c=IN IP4 <MyPublicIP>
t=0 0
m=audio 30454 RTP/SAVPF 102 9 8 0 3 18
a=rtpmap:102 SILK/8000
a=fmtp:102 useinbandfec=1; usedtx=0
a=rtpmap:9 G722/8000
a=rtpmap:8 PCMA/8000
a=rtpmap:0 PCMU/8000
a=rtpmap:3 GSM/8000
a=rtpmap:18 G729/8000
a=ptime:20
a=sendrecv
a=rtcp:30455
a=crypto:1 AES_CM_128_HMAC_SHA1_80
inline:5X2wgU83bIwrwkHSymjnw48SiMGa1n+eIl50yr99
a=setup:actpass
a=fingerprint:sha-1
11:76:2D:2A:F7:0D:5A:23:9D:F6:0C:E7:4C:DF:1E:CB:BF:5D:76:88
a=ice-ufrag:iglHZqry
a=ice-pwd:eyVXbFNoviYYlu5uuRZlnixFwQ
a=candidate:cE0FGbXWIfwj6OGD 1 UDP 2130706431 <MyPublicIP> 30454 typ host
a=candidate:cE0FGbXWIfwj6OGD 2 UDP 2130706430 <MyPublicIP> 30455 typ host
m=audio 30484 RTP/SAVPF 102 9 8 0 3 18
a=rtpmap:102 SILK/8000
a=fmtp:102 useinbandfec=1; usedtx=0
a=rtpmap:9 G722/8000
a=rtpmap:8 PCMA/8000
a=rtpmap:0 PCMU/8000
a=rtpmap:3 GSM/8000
a=rtpmap:18 G729/8000
a=ptime:20
a=sendrecv
a=rtcp:30485
a=crypto:1 AES_CM_128_HMAC_SHA1_80
inline:ozDJ3G/wmaYedQbcTafbhaTIt6raIJa6ugrLUC99
a=setup:actpass
a=fingerprint:sha-1
11:76:2D:2A:F7:0D:5A:23:9D:F6:0C:E7:4C:DF:1E:CB:BF:5D:76:88
a=ice-ufrag:gM1viWqu
a=ice-pwd:zdS4tP1Mj4hLTe6huBWYrhtI1s
a=candidate:cE0FGbXWIfwj6OGD 1 UDP 2130706431 <MyPublicIP> 30484 typ host
a=candidate:cE0FGbXWIfwj6OGD 2 UDP 2130706430 <MyPublicIP> 30485 typ host
My setup :
incoming call -> MediaServer (freeswitch) -> SIP & Websocket proxy
(Kamailio + Rtpengine) -> webRTC clients
Is there anybody who could help me with this issue or explain me how to
remove the second line of m=audio RTP/SAVPF ?
--
Jan
Hello,
We are using pua-dialoginfo and presence modules for BLF.
When the first call comes in or goes out, every thing works great; however,
a second call to the same key/extension (e.g a call waiting which is not
answered) would turn off the key even though the original call/dialog is
still active.
Has anyone had experienced such problem? How can I fix this?
Any constructive suggestion would be appreciated.
Regards,
Ali Pey
dear all
i'm using kamailio 4.4.1 and i'm not seeing nh_reload_rtpp as a fifo
exported function
i'm trying to setup the dburl for the rtpengine sets using db_text but
when i do kamctl fifo which i see all the commands of the module except
the reload_rtpp
nh_enable_rtpp
nh_show_rtpp
nh_ping_rtpp
nh_show_hash_total
/usr/local/kamailio/sbin/kamctl fifo nh_reload_rtpp
500 command 'nh_reload_rtpp' not available
these are the config lines i have for that
#!define DBURLRTP "text:///var/log/rtpengine_conf"
loadmodule "db_text.so"
modparam("rtpengine", "db_url", DBURLRTP)
modparam("rtpengine", "table_name", "rtpproxy")
and the database is like
root@test:/var/log/rtpengine_conf# ls
rtpproxy version
root@test:/var/log/rtpengine_conf# cat version
table_name(str) table_version(int)
rtpproxy:1
root@test:/var/log/rtpengine_conf# cat rtpproxy
set_id(int) url(str) weight(int) disable(int)
1:udp\:x.x.x.x\:x:100:0
2:udp\:x.x.x.x\:x:100:0
is there something i missed here? do i must use mysql or postgres
drivers for this? (i use a test machine to old to install a driver of
those with kam 4.4 maybe?)
best regards
david
Hello Daniel
i had the call rate for some minutes actually.
maybe 4-5min and all the time the behaviour was like that. it's true
that in that kamailio there wa no any other traffic, only those test
calls i set with a sipp to simulate the load
best regards
david
Hello,
for how long you got this 20req/second?
The module is not counting like continuous time, but more like traffic
on each second, iirc.
Cheers,
Daniel
On 23/05/16 11:24, david escartin wrote:
> Hello all
>
> i have a quick question about using module ratelimit on kamailio 4.2
> or 4.4.
> on tests made before running it on production, i set the RED
> algorithm with a limit of 10 calls per second, but i saw the call
> limitation was not done until we got more than 20 calls per second or
> so on the queue. we didnt get any INVITE discarded.
> after surpassing around 20 INV/sec, it's true the algorithm starts to
> work fine and only 10INV/sec are passed more or less
>
> on production i have no easy way to check this behaviour, besides it's
> a feature barely used.
>
> so i only want to know if this behaviour is known, or if i'm missing
> soemthing
>
> my config is like this
>
> modparam("ratelimit", "timer_interval", 5)
> modparam("ratelimit", "queue", "4:INVITE")
> modparam("ratelimit", "pipe", "4:RED:10")
>
> route[RATE_LIMIT] {
> $var(limitation)="4";
> if (!rl_check("$var(limitation)")) {
> send_reply("505","Limiting");
> xlog("L_INFO","Call $ci / Call-ID $ci: ratelimit was reached,
> rejecting with 505\n");
> route(CLEAR);
> exit;
> }
> }
>
>
> thanks a lot and regards
> david escartin
>
>
>
> _______________________________________________
> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
> sr-users at lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
--
Daniel-Constantin Mierla
http://www.asipto.com - http://www.kamailio.orghttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Hello Kamailio users,
I'm working on a sip infrastructure, with opensource technologies, and of
course Kamailio will run on my new infrastructure :)
Kamailio will be my central routing engine, between SBC (running on
asterisk), and somes boxes (asterisk too) where customers will be
registred. Please find below my future infrastructure:
---------- --------------
-------- ----------------
| SBC1 | - - - - - | - - - -| kamailio1 |- - - - - | - - - - - | CB1
|-----------------| SIPPHONE |
---------- | -------------- |
-------- ----------------
|
|
---------- | |
--------
| SBC2 | - - - - - | | - - - - - | CB2
|
---------- | |
--------
| -------------- |
---------- | - - - -| kamailio2 |- - - - -| --------
| SBCn |- - - - - | -------------- | - - - - - | CBn |
---------- | |
--------
Kamailio high availabality will be done with loopback address and BGP in
two datacenters. SIP carriers will be connected to SBC and customer to one
CB.
I have made some tests with kamailio, and i use LCR to route call from CB
to SBC, using wheigt and prorities to send call to carriers.
Then, when a call arrived from carriers, i use DNS NATPER lookup to find
the right CB box.
CB HA is done using DNS SRV. So user is registred at one location, and only
one. No location sharing beetwen boxes.
I have some question concerning my architecture, and technologies used to
routing call.
Is LCR is the best option for routing call between multiple carriers ? DNS
NAPTER is a modern solution to find an endpoint on this architecture ?
Existing a better way to route call with kamailio, for exemple rt-json and
curl ?
In my point of view, kamailio should be a routing engine, using internal
technologies, to maximise performances. Less work with kamailio and
external app, best performances in sip routing. Is it the good vision ?
Thanks in advance.
Hi there,
I'm looking for a solution to route calls based on prefix, my routing table
might reach a long list of prefixs.
I looked to carrierroute and it seems to be nice for this purpose but i
have some doubts to how to handle correctly with carrierroute.
IfI understood correctly, if I set the parameter avoid_failed_destinations
to 1, can i use the same carrier id and domain on failure route to select
another destination that matches the prefix?
Can I have the same carrier ID with diferent domains sharing the same
prefixs but mapping it to different destinations?
Thank you for your help.
Best regards
--
José Seabra
