Hi Gents,
It turned out that it ran with a wrong config and it was bounded to 0.0.0.0.
In this case the source IP is based on the routing decision, I can
reproduce it.
I think Kamailio should be smarter in this case.
Thanks anyway.
Cheers,
Mitya
On 06/26/2015 07:22 PM, Mihaly Zachar wrote
Hi Gents,
We have a redirect server based on Kamailio (4.2.3).
The server has multiple IPs, but Kamailio is bound to only one address:
http://view.xscreenshot.com/9dfdf449a387bdc6f9a719eae3620416
*Sometimes* it answers back from a wrong IP address, here is the proof:
http://view.xscreenshot.c
On 03/31/15 11:15, Olle E. Johansson wrote:
>
> On 31 Mar 2015, at 11:11, Mihaly Zachar wrote:
>
>> Hi All,
>>
>>
>> As far as I can see if a child dies due to segmentation fault Kamailio
>> stops all the childs and it exits.
>>
>> Is there
Hi All,
As far as I can see if a child dies due to segmentation fault Kamailio
stops all the childs and it exits.
Is there a possibility to change this behaviour to a kind of "respawn
the child" way ?
What do you think is the best way to keep Kamailio running on Ubuntu ?
On RHEL ?
Thanks,
Mis
Hi Daniel,
The patch was created on 4.1.4, sorry for that, but the bug still exists
at 4.2.3 (d76f8e3) and df86f2a as well.
Cheers,
Misi
On 03/18/15 17:45, Daniel-Constantin Mierla wrote:
> Hello,
>
> what version are you using?
>
> Cheers,
> Daniel
>
> On 18/03/15 16
Hi Gents,
Please consider applying this patch:
--- ./modules/sqlops/sql_api.c.orig 2015-03-18 16:33:56.994320505 +0100
+++ ./modules/sqlops/sql_api.c 2015-03-18 16:34:11.254419341 +0100
@@ -743,7 +743,7 @@
LM_ERR("row index out of bounds [%d/%d]\n", i, res->nrows);
Hi Gents,
We have created a small redirect server using the Lua API to get the
destinations.
Lets say that it takes some time while the script generates the answer,
during this time we got a CANCEL.
How we can handle this properly ? As far as I could see the TM module
always sends back the answer
