[Voiceglue] A CRITICAL entry
James Green
james.green at mjog.com
Fri Jun 5 11:10:28 UTC 2009
Hi Doug,
Asterisk logs for this call uploaded: http://pastebin.com/m407b4b91
I suspect we might not have full AGI logging switched on...
Perhaps we might short-cut a little here. We are trying to understand precisely what we can let Asterisk do (reliably) versus what we must capture and perform ourselves.
As a best practice, should VoiceXML applications capture disconnects if we do our app clean-up in the h section of the dial plan?
We took ours out (under the mistaken impression that should the callee hang up, VoiceXML should be destroyed) and instead performed clean-up in the hangup section of the dial plan (sounds logical!), it appeared to help, then we started seeing mass hangs of the VoiceGlue process under load at random, with DeadAGI listed for each channel.
We now capture disconnects (VoiceXML) again and exit, which seems to have solved that issue in the majority of cases. Really knowing what is expected of us (in reality) would be useful.
As a further best practice, should we ALWAYS have a dialplan 'h' (hangup) that performs application clean-up (database work) regardless of whether the callee or us (the caller) disconnected, or should this not be necessary?
Thanks,
James
-----Original Message-----
From: voiceglue-bounces at voiceglue.org [mailto:voiceglue-bounces at voiceglue.org] On Behalf Of Doug Campbell
Sent: 04 June 2009 17:58
To: General discussion about voiceglue
Subject: Re: [Voiceglue] A CRITICAL entry
> Further logs (same environment as before, but limited to five
> concurrent calls maximum): http://pastebin.com/m27917c43
13:32:16:119 DBUG PHONGLUE testvoic callid=[370] snd GETDIGIT timeout=7000 to AGI client on fh="::FASTAGI370" at host=127.0.0.1 callid=[370]
13:32:16:417 DBUG PHONGLUE testvoic callid=[370] AGI result=-1
The top line says that the phoneglue component of vliceglue just sent the AGI command "WAIT FOR DIGIT 7000" to Asterisk.
The bottom line says that Asterisk returned -1, a channel failure.
Such a failure gets passed to OpenVXI, which then reports a failure of the "VXIrecInterface" component, but it's just passing along the Asterisk error.
> Is there a catch within the VoiceXML that might capture these critical
> events? I'm really looking to make a call to a script to clean up as
> the disconnect occurs.
VXML does provide disconnect events, but the log provided here does not show a disconnect, it shows an Asterisk channel failure. If it were a normal Asterisk disconnect, the AGI connection would have been taken down by Asterisk. Now, it is possible that Asterisk is either being slow or sloppy with the disconnect handling.
I recommend checking the Asterisk logs to see why Asterisk is unhappy.
> P.S. I'm guessing there's no bugs database as yet?
Alas, you are correct.
Doug Campbell
_______________________________________________
Voiceglue mailing list
Voiceglue at voiceglue.org
http://www.voiceglue.org/mailman/listinfo/voiceglue
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.339 / Virus Database: 270.12.53/2154 - Release Date: 06/04/09 05:53:00
No virus found in this outgoing message.
Checked by AVG - www.avg.com
Version: 8.5.339 / Virus Database: 270.12.53/2155 - Release Date: 06/04/09 17:55:00
-------------- next part --------------
No virus found in this outgoing message.
Checked by AVG - www.avg.com
Version: 8.5.339 / Virus Database: 270.12.53/2156 - Release Date: 06/05/09 06:24:00
More information about the Voiceglue
mailing list