[asterisk-dev] Architectural problem with FSK callerid on analog lines

[Sebastian]

Hi,

I've seen in some places Argentina for example, that callerid is not 
updated from call to call. The exact scenario is like this:

1) Incoming call
2) not answered goes to a call queue
3) is ringing on an agent
4) the caller hangs up, but there's a new call coming (this is easy to 
see on busy systems or testing directly), the telco has no type of 
signalling to let the system knows there's a new call (in some countries 
you can see a polarity reversal that does the trick), but if not the 
next call hijacks the previous call thread, there's no way to know for 
ringtimeout.

there's an open case on digium: #00223214

but I think this should has some discussion, I'm not seen my mails pass 
to the mailing list lately.

What do you think about checking for fsk between rings??? is the only 
thing I can think of to at least to update the callerid although the 
call hijack the thread of the unattended call.
That also could solve any kind of callerid that is sent after the second 
ring or later if that exists in some country.


please confirm that my mails are getting to the mailing list, because 
I'm not seen them pass trough.

Thanks