[Asterisk-Users] Asterisk 1.0 & Zaptel 1.0 -- False Hangup Disaster

[Benjamin on Asterisk Mailing Lists]

On Sat, 25 Sep 2004 12:41:03 -0400, Andrew Kohlsmith
<akohlsmith-asterisk at benshaw.com> wrote:
> Fair enough, I saw that you'd written "tried every option" but a lot of people
> don't actually mean that.  :-)

:-)

> > Lines are provided by NTT. The driver (wcfxo.o) has been
> > built with "#define JAPAN" uncommented. Before the present 1.0
> > release, this has usually reduced false hangups.
> 
> Hmm okay so it is a known bug then;

Not sure whether it would have been considered a bug before. If you
look in the wcfxo code, the JAPAN define seems to simply set some
values differently, ie offhook-debounce or whatever it's called.

> have you done any hunting around on the
> bugtracker or bothered a bug marshall?

I have put it on the bugtracker, but Mark has declared it resolved for
technicality reasons I don't fully understand, which is not to say
that I am criticising it. I was lucky enough to catch Mark on the chat
and we talked about how to approach this. He told me that 1.0 was a
snapshot of last Thursday's or Friday's CVS and that I should find out
exactly when between 20 Aug and 1.0 this broke. So I will have a bit
of testing to do over the next days.

> possibly where I'd expect to sit and hang for a while

I had hoped to use 1.0 for an upcoming customer deployment because I
thought it was a decendent of RC2 and there was a feature freeze, but
now that I know it's just a very recent CVS I am not so sure I can
dive in head over heels, so I will use the CVS of May 1st which has
proven to be rather stable and with no or few surprises.

But I'll dedicate a machine or two to 1.0 testing.

> > Fair enough. But then again, why not have an option that disables
> > hangup detection until a call actually goes to voicemail and leave it
> > disabled if it doesn't?!
> 
> Becuase it's a workaround and doesn't actually address the problem?  In your
> case it might be a valid solution though; I wonder how hard it'd be to
> actually hack in?

Well, that's what I was wondering about. Workaround or not, if it
makes a huge difference for customers who would otherwise shun
Asterisk, then it shouldn't be too much of a religious concern.
Besides, I was suggesting this as an *option* that would by default be
disabled, so it wouldn't make any difference for those who have no
false hangups.

Anyway, we'll first have to find the culprit.

rgds
benjk

-- 
Sunrise Telephone Systems, 9F Shibuya Daikyo Bldg., 1-13-5 Shibuya,
Tokyo, Japan.

NB: Spam filters in place. Messages unrelated to the * mailing lists
may get trashed.