I reproduced this error with SipXecs 3.11.12-014925 and created issue XECS-2408.
Logfiles with "INFO"-loglevel have been attached.

This is the first issue that i created. I hope i did everything right.

René

> On Wed, 2009-03-25 at 13:25 +0100, Rene Pankratz wrote:
>   
>> Hello,
>>
>> We figured out a serious issue, that causes the sipXproxy process to 
>> take 100% CPU-usage and and breaks down the whole SipXecs.
>> We can reproduce this behaviour on SipX 3.10.2, SipX 3.10.3 and 
>> 3.11.12-014925.
>>
>> The following steps lead to this problem:
>> Create a hunt-group (e.g.extension 800) in SipX that consists of 3 
>> telephones (e.g. 801, 802, 803). All phones shall ring at the same time, 
>> don't allow call forwarding.
>> Modify the call forwardings of the hunt-group members via Webinterface, 
>> so that the other phones ring if user is busy or activated DND.
>>
>>    Extension 801:
>>       Forward to 802 if no response and ring 803 at the same time.
>>
>>    Extension 802:
>>       Forward to 803 if no response and ring 801 at the same time.
>>
>>    Extension 803:
>>       Forward to 801 if no response and ring 802 at the same time.
>>
>> With this setting SipX soon starts to act in an unexpected way. 
>> Unfortunately some phones seemed to ring forever and soon the whole SipX 
>> broke down.
>> Background is that the members of the hunt group shall forward their 
>> calls to the others when they leave their desk. We expected this could 
>> be done by configuring SipX in the way described above and activating 
>> DND when leaving the desk.
>>
>> Setting the call forwarding of the users to the hunt-group-extension 
>> causes the same beahviour.
>>
>> Any suggestions to solve this problem? We want to avoid setting call 
>> forwardings on the phones.
>>     
>
> I'd be very interested in seeing this reproduced, especially on either
> 3.10.3 or any recent 3.11.  In 3.10.3 we added loop detection that
> should have prevented this...
>
> If you can, set your logging to INFO and configure this, then stop the
> sipxpbx, delete your logs (/var/log/sipxpbx/*), restart the services,
> make this call and let it fail, then take a snapshot.  Open an issue in
> the tracker and attach the snapshot.  If it's too large to attach,
> contact me and we'll find a way to get it, but if you delete the logs
> before you start this it probably won't be.
>
>
>
>
>   

_______________________________________________
sipx-users mailing list
[email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-users
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-users

Reply via email to