[Asterisk-Users] X100P troubles?

[Rich Adamson]

> > About a year and a half ago when I was running a couple of x100p's
> > there was an issue associated with disconnecting the pstn line from
> > the card. If I recall correctly, if the pstn line was removed for
> > more then a second or so (a couple of times), the card would go into
> > some unknown state and it had to be restarted. The comments/fix at the
> > time were oriented around "don't do that" instead of fixing the code
> > to recognizing it.
> >
> > I don't have a clue whether that is still an issue with the x100p
> > code or not.
> 
> Humm, when I moved the jacks around (trying to get the 'bad' card out  
> of the loop) the machine hung.

That sounds very familiar. :(

> > You might try using a regular old voltmeter on the tip/ring of the
> > line going to the Toshiba and watch what happens at the end of a
> > VM call. It is possible the Toshiba is opening/disconnecting the
> > line for some rather lengthy period (disconnect supervision).
> 
> I fear you might be right.  I used zttool a few times while the  
> machine was in production.  At least once I saw the card which gets  
> the most use in a 'red' state and then switch back to OK when  
> activity (ringing) occured.
> 
> Sigh, looks like a hardware/driver issue that's been around for at  
> least a year and a half?!  Say it isn't so. :'(

As I recall, the driver for the x100p was called wcfxs (or something 
like that), and those driver functions were merged into wctdm about a 
year ago. Now, wcfxs is an alias for wctdm.

The chipset that was used on the x100p was the Silicon Labs 3012 and
3021, which are different part numbers then what is used on the TDM 
modules. You can download the spec sheets for those chips and try to
debug the wctdm code. There is no one at digium that will work on it
anymore.