[asterisk-r2] Channels get stuck in idle/seize

Moises Silva moises.silva at gmail.com
Thu Nov 5 09:28:32 CST 2009


It seems like some sort of bug. Which Asterisk version and patch are you
using again? I forgot.

On Mon, Nov 2, 2009 at 1:54 PM, Christian Cabrera <ccabrera at gmail.com>wrote:

> Moy,
>
> Here are the dumps:
>
> http://pastebin.com/m5aa3f5d9
> http://pastebin.com/m27d373e0
>
> In case I did something wrong with the script, let me know so that I can
> get the information again.
>
>
> Kind regards,
>
>   Ing. Christian Cabrera
>
>   Nextel (777) 113 8204 Id 72*15*24335
>
>
> On Thu, Oct 29, 2009 at 13:51, Moises Silva <moises.silva at gmail.com>wrote:
>
>> try downloading the svn release-1 in google code, then use the script at
>> debug/asterisk/astdumpr2.sh, execute that script whenever you have stuck
>> channels. The script must be executed 2 times with an interval of 2 seconds
>> between each execution and pastebin the results.
>>
>> On Thu, Oct 29, 2009 at 3:36 PM, Christian Cabrera <ccabrera at gmail.com>wrote:
>>
>>> Moy,
>>>
>>> About 2 months ago you helped me out with a problem that made the
>>> channels of an E1 get stuck in CLEAR FO mode. Now I'm having a similar
>>> problem, but with SEIZE.
>>>
>>> I just did a reinstall of a system using Elastix 1.5.2. I'm using the
>>> most up to date version of DAHDI (2.2.0.2) and Asterisk 1.4.26.2 patched
>>> with the code provided in your Google's code page. I'm also using Wanpipe
>>> 3.5.8.
>>>
>>> The problem I'm having is that after the channels change to Tx/Rx
>>> Idle/Seize, Asterisk tries to take the channels for an outgoing call, and
>>> then it fails, giving out this:
>>>
>>> [Oct 28 16:01:39] ERROR[8446] chan_dahdi.c: Chan 51 - Trying to dial out
>>> in a
>>> non-idle channel (cas=0x00)
>>> [Oct 28 16:01:39] ERROR[8446] chan_dahdi.c: unable to make new MFC/R2
>>> call!
>>>
>>> Eventually, the channels go back to idle/idle after about 60 to 120
>>> seconds, but due to the volume of calls being processed in this system, this
>>> isn't feasible, since too many errors occur (sometimes we get something as
>>> low as a 50% success ratio in outgoing calls). Incoming calls do not present
>>> this problem at all.
>>>
>>> The provisional solution I've created is a PHP script which recurrently
>>> sets the channels in idle/seize to block/seize (that way Asterisk wont try
>>> to use them and won't give out any errors). After a while, when the channels
>>> switch to block/idle, I change them to idle/idle, which allows Asterisk to
>>> do the calls.
>>>
>>> I'm using a Maxcom E1 in Mexico city.
>>>
>>> Any thoughts on how to solve this?
>>>
>>> Kind regards,
>>>
>>>   Ing. Christian Cabrera
>>>
>>>   Nextel (777) 113 8204 Id 72*15*24335
>>>
>>> _______________________________________________
>>> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>>>
>>> asterisk-r2 mailing list
>>> To UNSUBSCRIBE or update options visit:
>>>   http://lists.digium.com/mailman/listinfo/asterisk-r2
>>>
>>
>>
>>
>> --
>> Moises Silva
>> Software Developer
>> Sangoma Technologies Inc. | 50 McIntosh Drive, Suite 120, Markham ON L3R
>> 9T3 Canada
>> t. 1 905 474 1990 x 128 | e. moy at sangoma.com
>>
>> _______________________________________________
>> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>>
>> asterisk-r2 mailing list
>> To UNSUBSCRIBE or update options visit:
>>   http://lists.digium.com/mailman/listinfo/asterisk-r2
>>
>
>
> _______________________________________________
> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>
> asterisk-r2 mailing list
> To UNSUBSCRIBE or update options visit:
>   http://lists.digium.com/mailman/listinfo/asterisk-r2
>



-- 
Moises Silva
Software Developer
Sangoma Technologies Inc. | 50 McIntosh Drive, Suite 120, Markham ON L3R 9T3
Canada
t. 1 905 474 1990 x 128 | e. moy at sangoma.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.digium.com/pipermail/asterisk-r2/attachments/20091105/d6ad982d/attachment.htm 


More information about the asterisk-r2 mailing list