Is it normal for Kamailio to segfault on a duplicate key in the usrloc DB?
Jun 15 18:15:23 vproxy2 /usr/sbin/kamailio[13646]: ERROR: db_mysql
[km_dbase.c:122]: db_mysql_submit_query(): driver error on query: Duplicate
entry 'uloc-557a96c1-3577-4e5f1' for key 'ruid_idx'
Jun 15 18:15:23 vproxy2
for 4.0.6. Only 4.0.7
is in the repo.
Anyway, I still believe the trouble is with writing to the DB.
Marc
On Tue, Jun 16, 2015 at 12:05 PM, Alex Balashov abalas...@evaristesys.com
wrote:
On 06/16/2015 12:04 PM, Marc Soda wrote:
Is it normal for Kamailio to segfault
No. :-) Can you
I'm running a new kamailio/rtpengine instance and, using webrtc, I can't
get audio to flow between the browser and rtpengine. Kamailio seems to be
handing it off properly.
Based on the rtpengine logs and a packet capture it doesn't look like DTLS
is being negotiated properly:
I wound up upgrading rtpengine and that resolved this issue. I ran into
something new, but I opened an issue for it (here:
https://github.com/sipwise/rtpengine/issues/92 if anyone is interested).
Sorry for the noise.
On Wed, Mar 25, 2015 at 1:46 PM, Marc Soda ms...@coredial.com wrote:
I'm
I'm wondering if anyone has any experience load testing with Kamailio and
rtpengine using WebRTC. I've had success load testing just Kamailio with
sipp, but now I'd like to add the media piece and, if possible, do it via
WebRTC.
Anyone have any thoughts?
Thanks,
Marc
How does Kamailio load balance traffic to rtpengine? Is it load based,
round robin, etc? The module makes mention of this but I don't see how it
works. Also, it talks about weighting the proxies. How is that
accomplished?
Thanks,
Marc
___
SIP
at 8:58 PM, Marc Soda ms...@coredial.com wrote:
We are in the middle of designing a similar solution with Kamailio and
rtpengine and after some initial problems things are going really well. I
can tell you that we ended up going with SIPjs over JSSip and it handled a
lot of the weird browser
We are in the middle of designing a similar solution with Kamailio and
rtpengine and after some initial problems things are going really well. I
can tell you that we ended up going with SIPjs over JSSip and it handled a
lot of the weird browser specific issues we were having.
I'm not sure about
in the modparam() connection string, as you do with other DB
modules. Maybe calling something like this out in the documentation would
help clear that up.
On Thu, Jan 29, 2015 at 3:48 PM, Marc Soda ms...@coredial.com wrote:
Ah, so you should be able to do something like:
redis_cmd(localredis, AUTH $var
The only way it will work right now is to not use a password:
modparam(ndb_redis, server, name=localredis;addr=localhost;port=6379)
I've been wanting to look at contributing support at that, but no time...
On Thu, Jan 29, 2015 at 10:16 AM, Yuriy Gorlichenko ovoshl...@gmail.com
wrote:
Hello. I
after connecting
and can be done from config -- that's on a very quick search, not sure if
something has changed with the hiredis api meanwhile:
- https://github.com/redis/hiredis/issues/56
Cheers,
Daniel
On 29/01/15 17:20, Marc Soda wrote:
The only way it will work right now
On Fri, Dec 19, 2014 at 12:31 PM, Richard Fuchs rfu...@sipwise.com wrote:
You've caught the same thing as Juha did just earlier, Firefox is doing
something new called Trickle ICE, which at the moment breaks
communications with endpoints not supporting it (such as rtpengine).
I figured out
Does anyone know how a can get stats from rtpengine? I see the $rtpstat
pseudo variable in Kamailio, but from the documentation it looks like that
will only give me stats on a particular call. I'm looking for overall
stats like concurrent calls, bandwidth, etc...
Thanks,
Marc
That's how I ended up going. It's working now. Thanks.
On Thu, Dec 18, 2014 at 4:11 PM, James Cloos cl...@jhcloos.com wrote:
MS == Marc Soda ms...@coredial.com writes:
MS I'm having a problem reassembling UDP packets on my Asterisk servers
after
MS passing through Kamailio
You could
I'm trying to use Kamailio and rtpengine as a webrtc gateway. I'm not
getting audio back to my browser. From a packet capture I can see media
from the browser to rtpengine, and then bi-directional RTP back and forth
from my asterisk server, but rtpengine is not sending the media on to the
On Fri, Dec 19, 2014 at 10:47 AM, Marc Soda ms...@coredial.com wrote:
I'm trying to use Kamailio and rtpengine as a webrtc gateway. I'm not
getting audio back to my browser. From a packet capture I can see media
from the browser to rtpengine, and then bi-directional RTP back and forth
from
Thanks for the response. You're right, the media stream is making it all
the way back to my PC, I just don't hear anything. And yes, my speakers
are turned up...
I'm not sure what to try next...
On Fri, Dec 19, 2014 at 12:31 PM, Richard Fuchs rfu...@sipwise.com wrote:
On 12/19/14 10:47, Marc
, 2014 at 5:06 AM, Daniel-Constantin Mierla mico...@gmail.com
wrote:
On 18/12/14 02:58, Marc Soda wrote:
So gzcompress is no good with Asterisk then? Is that meant to be used
only with another Kamailio proxy?
Apparently Apple Facetime is using this kind of compression (as it was
reported
I'm having a problem reassembling UDP packets on my Asterisk servers after
passing through Kamailio (it appears to me an OS level issue, nothing to do
with Kamailio). Is there a way, with Kamailio, to limit the size of a SIP
message? I know I can just start removing headers, but that doesn't
-Constantin Mierla mico...@gmail.com
wrote:
On 17/12/14 23:20, Alex Balashov wrote:
On 12/17/2014 05:14 PM, Marc Soda wrote:
I'm having a problem reassembling UDP packets on my Asterisk servers
after passing through Kamailio (it appears to me an OS level issue,
nothing to do with Kamailio
Can someone recommend a way to extract the user part of a Contact header
URI?
Right now I'm just trying to pass it to xlog(). I have tried:
- xlog(L_NOTICE, $(ct{uri.user}));
This gives me an empty string.
- xlog(L_NOTICE, @contact.uri.user);
This gives me an error starting Kamailio: function
Finally! This seems to have done it:
xlog(L_NOTICE, $sel(contact.uri.user));
Sorry for the noob question.
On Mon, Dec 8, 2014 at 11:14 AM, Marc Soda ms...@coredial.com wrote:
Can someone recommend a way to extract the user part of a Contact header
URI?
Right now I'm just trying to pass
Hey all,
I have a Kamailio box functioning as a proxy. When some UAs send an INVITE
the SDP has their private IP (c=IN IP4 10.x.x.x). Kamailio is passing this
on without changing it to the proper public IP. What's the best way to
rewrite that? I'm using a config based on the default one.
I have a Kamailio server forwarding REGISTERs to an Asterisk box, similar
to the way Daniel shows here:
http://kb.asipto.com/asterisk:realtime:kamailio-4.0.x-asterisk-11.3.0-astdb
Can I force Kamailio to append rport=5060 to the topmost Via header, prior
to forwarding the REGISTER to Asterisk?
are sent to 1.1.1.1:59738 and not 1.1.1.1:5060. I need
them to go to 1.1.1.1:5060.
On Fri, Mar 28, 2014 at 2:28 PM, Marc Soda ms...@coredial.com wrote:
I have a Kamailio server forwarding REGISTERs to an Asterisk box, similar
to the way Daniel shows here:
http://kb.asipto.com
to is generating a new REGISTER from Kamailio to
Asterisk, putting in it the Contact header with the address of kamailio. So
Asterisk should send the INVITE to kamailio.
Cheers,
Daniel
On 28/03/14 19:37, Marc Soda wrote:
Basically, I'm trying to get Asterisk to send all future INVITEs to
Kamailio
I'm trying to handle a redirect with get_redirects(). It seems that
Kamailio is ignoring the maddr param on the contact header. Is there a way
to force maddr to be used?
The Contact header on the 302 looks like this:
sip:2404441...@domain.com:5060
I found t_on_branch_failure() in 4.1. Would that be the way to handle this?
On Thu, Mar 13, 2014 at 1:36 PM, Marc Soda ms...@coredial.com wrote:
Can someone tell me how to handle a 401 from a 302 redirect? I am
attempting to register with the uac module. Normally, I set a failure
route
Can someone tell me how to handle a 401 from a 302 redirect? I am
attempting to register with the uac module. Normally, I set a failure
route for the 401 and call auc_auth(). In this case, I am receiving a 302,
sending the register with get_redirects(), then getting a 401. However,
Kamailio
I have Kamailio setup as a proxy in front of a backend server (Asterisk).
When I make a call through the proxy, the Contact header in the 200 OK
that is returned to the client has the IP of the backend server in it.
Thus, the client is sending it's ACK directly to the backend server.
Is there a
PCMA/8000
a=rtpmap:101 telephone-event/8000
a=fmtp:101 0-16
a=silenceSupp:off - - - -
a=ptime:20
a=sendrecv
I don't think 3.3.3.3 should show up anywhere, it should be rewritten to
2.2.2.2.
On Wed, Mar 5, 2014 at 1:34 PM, Olle E. Johansson o...@edvina.net wrote:
On 05 Mar 2014, at 18:30, Marc
, but containing a route header pointing to the Kamailio IP.
Kamailio will loose_route() this request and send it to the backend server
as expected.
Regards,
On Wed, Mar 5, 2014 at 3:53 PM, Marc Soda ms...@coredial.com wrote:
Thanks Olle. I am calling record_record() on the initial INVITE
Hey all,
I have a pretty general SIP question that I'm hoping some of you can
shed some light on. I hope this ok for the list.
I am setting up a SIP proxy with Kamailio. The backend server
(Asterisk in my case) requires authentication. Is it standard/best
practice to require a proxy to
I forget to mention, the nat device is in front of the Kamailio servers,
not the endpoints.
On Fri, Feb 28, 2014 at 6:22 PM, Marc Soda ms...@coredial.com wrote:
I have a Kamailio server setup which is registers to a back end server on
behalf of endpoints. The endpoints can register
So I've found out that NAT has nothing to do with it. The bit about things
working when the NAT device is removed was wrong.
So my question becomes: Why would Kamailio ignore a 401 rather sending it
to a failure route?
Thanks in advance,
Marc
On Mon, Mar 3, 2014 at 9:10 AM, Marc Soda ms
, 2014 at 11:52 AM, Daniel-Constantin Mierla mico...@gmail.com
wrote:
Are you sure you have set t_on_failure() for the respective transaction?
Cheers,
Daniel
On 03/03/14 17:44, Marc Soda wrote:
So I've found out that NAT has nothing to do with it. The bit about
things working when the NAT
On Mon, Mar 3, 2014 at 12:24 PM, Daniel-Constantin Mierla mico...@gmail.com
wrote:
On 03/03/14 18:08, Marc Soda wrote:
I resolved the issue, but I not quite sure why is worked. Rather than
sending the REGISTER with t_reply()
t_reply() is not sending REGISTER anywhere, it is sending
I have a Kamailio server setup which is registers to a back end server on
behalf of endpoints. The endpoints can register to Kamailio but Kamailio
is failing to register to the server when I put a NAT device in front of
it. Without the NAT device it works fine.
The problem is the 401 that comes
Also, I tried setting the Contact header in onsend_route, but it seems that
it's overwritten after that. Perhaps I need to modify it on the second
Kamailio server? Is it possible to change it before it's sent on the first
server?
On Wed, Sep 4, 2013 at 8:53 AM, Marc Soda ms...@coredial.com
once it is
received on 3.3.3.3 is user@2.2.2.2. I think it should be user@1.1.1.1.
Where's the best place to change this before it is sent?
--
Marc Soda, Sr. Systems Engineer
*CoreDial, LLC* | www.coredial.com
1787 Sentry Parkway West, Building 16, Suite 100, Blue Bell, PA 19422
Office: (215) 297
On Thu, Aug 29, 2013 at 3:53 AM, Daniel-Constantin Mierla mico...@gmail.com
wrote:
what device is at 701? The 200ok receved from it has the contact address
with the IP of kamailio:
SFLPhone
It seems there is a NAT between your kamailio and 701, as kamailio adds
alias parameter to
;
}
}
Thanks a lot Daniel!
On Thu, Aug 29, 2013 at 10:58 AM, Marc Soda ms...@coredial.com wrote:
On Thu, Aug 29, 2013 at 3:53 AM, Daniel-Constantin Mierla
mico...@gmail.com wrote:
what device is at 701? The 200ok receved from it has the contact address
with the IP
I think I found my missing ACKs! Can anyone tell me why they work be
being sent to the loopback interface? The destination address is
still the external (eth0) IP.
--
Marc Soda, Sr. Systems Engineer
*CoreDial, LLC* | www.coredial.com
1787 Sentry Parkway West, Building 16, Suite 100, Blue
Thanks, I appreciate it.
In this setup the there are 2 endpoints (700 and 701) peered up to an
Asterisk server (172.16.60.6) via a Kamailio proxy (172.16.60.20). 700
(172.16.60.28) is calling 701 (172.16.3.65). When 701 answers the OK is
sent to the proxy and then to Asterisk. Asterisk is then
I trying to debug why Kamailio is dropping certain ACKs. I have debug
level set to 4 and cfgtrace on. The last thing I see in the log for my
missing ACKs is:
Aug 26 12:22:29 eng-reg1 kamailio: ERROR: *** cfgtrace:
c=[/etc/kamailio/kamailio.cfg] l=789 a=16 n=if
Aug 26 12:22:29 eng-reg1 kamailio:
(Sorry for the double post)
I trying to debug why Kamailio is dropping certain ACKs. I have debug
level set to 4 and cfgtrace on. The last thing I see in the log for my
missing ACKs is:
Aug 26 12:22:29 eng-reg1 kamailio: ERROR: *** cfgtrace:
c=[/etc/kamailio/kamailio.cfg] l=789 a=16 n=if
Aug
-router.org/cgi-bin/mailman/listinfo/sr-users
--
Marc Soda, Sr. Systems Engineer
*CoreDial, LLC* | www.coredial.com
1787 Sentry Parkway West, Building 16, Suite 100, Blue Bell, PA 19422
Office: (215) 297-4400 x203 | Fax: (215) 297-4401 | Email:
ms...@coredial.com
- - - - -
The information
I think I found my missing ACKs! Can anyone tell me why they work be being
sent to the loopback interface? The destination address is still the
external (eth0) IP.
On Mon, Aug 26, 2013 at 3:36 PM, Marc Soda ms...@coredial.com wrote:
It's checking the source of the current message
Take a look at where route(frompstn) is being called. It's probably in a
'if' statement, meaning that if the source address is the pstn.gw_ip, then
return TRUE. Returning -1 is like saying, FALSE.
___
SIP Express Router (SER) and Kamailio (OpenSER) -
49 matches
Mail list logo
