[asterisk-dev] remote call supervision

[Tzafrir Cohen]

On Mon, Mar 03, 2008 at 08:09:52PM -0700, Bob wrote:
> Currently there is no difference between loop start and kewl start in the
> receive direction in zaptel. In the transmit direction there is a provision
> for removing battery for ZT_KEWL_TIME when the near end hangs up. There is a
> recent hack patch that has made its way into the trunk branches that alarm
> the span or channel when there is no battery. This patch is OK, but it
> shouldn't be used to detect far end disconnect. 
> 
> IMHO, the proper way to handle the far end disconnect, and the battery alarm
> is to implement an RX_KEWL state into zaptel instead of hoping that the DAA
> interface will die off and report when it's battery is pulled away from the
> far end. Some don't die. Some take longer to use the charge already stored.

Fine. Patches are welcomed.

> 
> All of our equipment will hang up on its own when battery is removed for a
> specified period. We use the standard 0xf on the rx abcd bits to represent
> this, and everything works fine. The problem with zaptel is that it is
> waiting for its own fxo circuit to hang up instead of detecting the lack of
> battery and doing so itself. 

I noticed that you guys from Rhino, unlike most others on this list,
tend to use the bits signalling. I must say that I find it to make it
slightly more complicated to understand what you mean.

> 
> This requires modification to zaptel, and chan_zap. I'm asking that someone
> at the big D, just go ahead and do this instead of messing with all the
> small patches and headaches.

While I'm not from Digium, I figure that if you have any helpful patches
they would be appreciated.

-- 
               Tzafrir Cohen
icq#16849755              jabber:tzafrir.cohen at xorcom.com
+972-50-7952406           mailto:tzafrir.cohen at xorcom.com
http://www.xorcom.com  iax:guest at local.xorcom.com/tzafrir