Topic: Sail keeps disconnecting incoming calls

[gippsweb]

Hi Jeff

Since loading Sail and having it running for a few weeks now, we are constantly having the zap card hanging up incoming PSTN calls at random times.
To check the card was not at fault I put it back into it's old a@h home and it doesn't miss a beat.
Outgoing calls over the internal zap line do not seem to be affected.

the only things I'm seeing in the logs are;

SIP/5001-9051 answered Zap/4-1
Hungup 'Zap/1-2'
AGI Script selintra completed, returning 0
Timeout on Zap/4-1
CDR updated on Zap/4-1
Executing Hangup("Zap/4-1", "") in new stack
Spawn extension (internal-presets, t, 1) exited non-zero on 'Zap/4-1'
Executing Hangup("Zap/4-1", "") in new stack
Spawn extension (internal-presets, h, 1) exited non-zero on 'Zap/4-1'

And most of that makes little sense to me.....

Cheers

Mark

[SARK devs]

Hello Mark,

Your disconnect looks like a regular "far-end" hangup.  However, I am puzzled by the internal-preset context being driven by the near-end time out.   Can you tell me more about this call?  Was it a normal ibound call?  

You might also want to correctly set your card settings for Australia.  Add the following line to /etc/modprobe.conf

 options wctdm opermode=AUSTRALIA

once you've done this, reboot your system.

It may make a difference.

Kind Regards

Selintra

[gippsweb]

This only happens on incoming calls and only ones that come in on the ZAP channel.
If I pick up a phone b4 the zap card the call will continue without problems.
The other thing is I think the hangup is happening internally on the extensions and not to the outside line.

Why do I think this;

* sometimes if I hangup my phone straight away and wait a few seconds either my phone or the other extension will start ringing and the person will still be there. (they here only silence as if I had muted then, not music as if they were on hold).

* other times I can pick up the handset on the fax machine and continue the conversation, thus taking the software out of the equation.

Looks like I had opermode=Australia in the wrong place.
I had it where one would for aah2, which must use a newer version of asterisk than I am now.
But things have gone from bad to worse;
Now it loads the bits below correctly,
Jul 18 13:15:55 gw1 kernel: Freshmaker version: 73
Jul 18 13:15:55 gw1 kernel: Freshmaker passed register test
Jul 18 13:15:55 gw1 kernel: Boosting ringer on slot 1 (89V peak)
Jul 18 13:15:55 gw1 kernel: Module 0: Installed -- AUTO FXS/DPO
Jul 18 13:15:55 gw1 kernel: Module 1: Not installed
Jul 18 13:15:55 gw1 kernel: Module 2: Not installed
Jul 18 13:15:55 gw1 kernel: Module 3: Installed -- AUTO FXO (AUSTRALIA mode)
Jul 18 13:15:55 gw1 kernel: Found a Wildcard TDM: Wildcard TDM400P REV I (2 modules)
Jul 18 13:15:56 gw1 kernel: Registered tone zone 1 (Australia)

but when I do a zaptel show status I get;
Wildcard TDM400P REV I Board 1   UNCONFIGUR 0    0    0
ZTDUMMY/1 1                               UNCONFIGUR 0    0    0
and the card is not working, but the lights are on (not sure if anyones home).

[SARK devs]

Hi Mark,

This does look as tho' its a bug.  Can you report on the contribs bugs site please.

Include all your release levels (do rpm -q on each selintra package you have installed - including the asterisk releases).

Include listings for the following...

uname -r
/etc/modprobe.conf
/etc/zaptel.conf
/etc/asterisk/zapata.conf/
/proc/zaptel/1 & 2 (if 2 is present)

We may also need to ssh into your machine to have a look at what's happening and place a few calls to it.  If you are happy for us to do this, please allow ssh logins on your sme-server and send details to admin@selintra.com

Kind Regards

Selintra

[gippsweb]

Jeff

Bug being logged while I deal with customers, no problems with SSH access. We'll just need to work out a time as I've had a few problems with a few idiots trying to brute force it.