[OpenSIPS-Users] radius SQL accounting of failed calls

[alfredoc]

Hi. I've also applied the same patch on freeradius-server-2.1.3, and
apparently went well. No errors during patching and compilation, but I'm
still getting the following error in radius log:

Sun Feb  8 21:40:22 2009 : Error: rlm_radutmp: Logout entry for NAS OpenSIPS
port 5060 has wrong ID
Sun Feb  8 21:40:27 2009 : Error: rlm_radutmp: NAS OpenSIPS port 5060
unknown packet type 15)

I've included dictionary.opensips in /usr/local/etc/raddb/dictionary file:

$INCLUDE        /usr/local/share/freeradius/dictionary
$INCLUDE        /usr/local/etc/raddb/dictionary.opensips
 
The same for radiusclient dictionary.

In opensips.cfg, radiusclient config :
.
.
modparam("acc", "failed_transaction_flag", 1)
modparam("acc", "radius_config",     
"/usr/local/etc/radiusclient-ng/radiusclient.conf")


Running on Centos 5.2, opensips 1.4.2-tls, rediusclient-ng.0.5.6 


Maybe I'm not using the correct patch or dictionary files ??  Any help will
be appreciated.

Regards,

Alfredoc



Jeff Pyle wrote:
> 
> Hi Bogdan,
> 
> It is CDRTool/contrib/freeradius-brandinger/freeradius_20080103.patch
> within
> the CDRTool tarball.
> 
> 
> - Jeff
> 
> 
> 
> On 1/9/09 5:14 AM, "Bogdan-Andrei Iancu" <bogdan at voice-system.ro> wrote:
> 
>> Hi Jeff,
>> 
>> Could you point me the FREERADIUS patch you are talking about? just to
>> take a look and maybe push the discussion to a developer from Free
>> Radius.
>> 
>> Thanks and regards,
>> Bogdan
>> 
>> Jeff Pyle wrote:
>>> Hi Bogdan,
>>> 
>>> That makes sense to me.  In fact, that seemed to be the central point of
>>> the
>>> argument against using a STOP record for a failed call:  "You can't STOP
>>> what never STARTed in the first place."  Perhaps the argument from the
>>> other
>>> side is that one must take a unique identifier into consideration?  I
>>> cannot
>>> claim to understand the implications yet.  I started playing with radius
>>> accounting only last week.  :)
>>> 
>>> Thankfully, the patch buried within the contrib directory of CDRTool
>>> applied
>>> well against freeradius-server-2.1.3.  Failed calls from OpenSIPS now
>>> cause
>>> SQL records to be inserted.  Inspecting the inserted records, however, I
>>> don't see anything indicating whether it was a START, STOP, or FAILED at
>>> the
>>> radius level.  Perhaps the only indicator is the session time?  I'm
>>> still
>>> investigating.
>>> 
>>> 
>>> - Jeff
>>> 
>>> 
>>> 
>>> 
>>> 
>>> On 12/15/08 6:36 AM, "Bogdan-Andrei Iancu" <bogdan at voice-system.ro>
>>> wrote:
>>> 
>>>   
>>>> Hi Jeff,
>>>> 
>>>> OpenSIPS is still sending the FAILED values for the missed calls. From
>>>> ACC point of view, you have two cases - A) established calls
>>>> (START+STOP) and B) failed calls (FAILED).
>>>> 
>>>> If you use the STOP also for the failed calls, wouldn't be a confusion
>>>> in between the STOP of an ongoing call and the STOP of a failed call?
>>>> 
>>>> Regards,
>>>> Bogdan
>>>> 
>>>> Jeff Pyle wrote:
>>>>     
>>>>> Hello,
>>>>> 
>>>>> It seems there was a long thread some time back about using the
>>>>> so-called non-standard ³Acct-Status-Type = Failed² in the radius
>>>>> packet. The two proposed solutions seemed to be 1) use a Stop type
>>>>> instead of Failed, or 2) patch Freeradius.
>>>>> 
>>>>> Since I¹m still seeing the Failed type in the radius detail file, it
>>>>> appears OpenSIPS didn¹t change to Stop.
>>>>> 
>>>>> So, anyone know of a patch for current Freeradius sources?
>>>>> 
>>>>> 
>>>>> Thanks,
>>>>> Jeff
>>>>>       
>>> 
>>> 
>>>   
>> 
> 
> 
> _______________________________________________
> Users mailing list
> Users at lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
> 
> 

-- 
View this message in context: http://n2.nabble.com/radius-SQL-accounting-of-failed-calls-tp1649525p2295455.html
Sent from the OpenSIPS - Users mailing list archive at Nabble.com.