asterisk users - Dec 2004 - PRI/Zap premature dialing problem

The originating PRI system passes the entire dialed number in the d-channel
setup frame, thus the concept of a wait time for additional digits is
meaningless. Progressive digit gathering implies that the signalling is
occuring 'in-band' as would be the case with DTMF signalling on analog
lines.

You need to look in the Ascom and find the configuration table that lays out
the dialplan for the network. The PBX itself will use that information to
determine that for example, if a number begins with '4' then it is
'complete' after a total of 3 digits (eg. 411) or if it begins with 1,
then
it's complete after 10 digits (1XXXNNNXXXX). After the number is considered
'complete' it will build the setup frame and transmit it to the
terminating
system. 

Watch the output of 'pri debug span 1' when the originating system is
dialing different numbers to see this in action.

Hope that helps.
> -----Original Message-----
> From:	Jerry Glomph Black [SMTP:asterisk-users@glomph.com]
> Sent:	Monday, December 06, 2004 5:16 AM
> To:	Asterisk Users Mailing List - Non-Commercial Discussion
> Subject:	[Asterisk-Users] PRI/Zap premature dialing problem
> 
> I know this may be RTFM flamebait, but I've spent a lot of time on the
> wiki, and 
> googling, etc. with no resolution.
> 
> Closest I could get is the wiki page:
> http://www.voip-info.org/wiki-Asterisk+Extension+Matching
> on the topic "Match As You Go" Dialing.
> ---------
> here's the problem:
> 
> I have extension 224, and also extension 2246.
> 
> A call inbound from the PRI requesting 2246 is always cut off at the first
> 
> match, instantly, when people are dialing in from conventional extensions.
> 
> Thus, 2246 is not reachable.    Messing with parsing-priorities (via the
> use of include files) does not seem to help.
> 
> The conventional (Ascom) PBX is presenting the digits immediately to the
> Zap 
> channel, and the channel is not waiting -at all- to see if the dialing is 
> finished, it instantly jumps to the first match.  The above wiki page says
> it 
> should wait 3 seconds...  But it is not doing so.
> 
> Thanks,
> Dr G
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users@lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-users

Kris, thanks for the thoughful & helpful response!   This makes sense in the
same way that a dialplan on a SIP phone would behave.

But... If I  remove the 3-digit number (224) from the asterisk 
dialplan,  I have no problem dialing 2246 successfully.   If the Ascom were just
seeing a 3-digit dialplan match for 224, would the fourth digit still be 
transmitted along the PRI?   This is the basis of the puzzle, it's as if 
asterisk is feeding back some info about whether a complete number has been 
passed along.



On Mon, 6 Dec 2004, Kris Boutilier wrote:
> The originating PRI system passes the entire dialed number in the d-channel
> setup frame, thus the concept of a wait time for additional digits is
> meaningless. Progressive digit gathering implies that the signalling is
> occuring 'in-band' as would be the case with DTMF signalling on
analog
> lines.
>
> You need to look in the Ascom and find the configuration table that lays
out
> the dialplan for the network. The PBX itself will use that information to
> determine that for example, if a number begins with '4' then it is
> 'complete' after a total of 3 digits (eg. 411) or if it begins with
1, then
> it's complete after 10 digits (1XXXNNNXXXX). After the number is
considered
> 'complete' it will build the setup frame and transmit it to the
terminating
> system.
>
> Watch the output of 'pri debug span 1' when the originating system
is
> dialing different numbers to see this in action.
>
> Hope that helps.
>
>> -----Original Message-----
>> From:	Jerry Glomph Black [SMTP:asterisk-users@glomph.com]
>> Sent:	Monday, December 06, 2004 5:16 AM
>>
>> A call inbound from the PRI requesting 2246 is always cut off at the
first
>>
>> match, instantly, when people are dialing in from conventional
extensions.
>>

On Mon, 6 Dec 2004, Kris Boutilier wrote:
> The originating PRI system passes the entire dialed number in the d-channel
> setup frame, thus the concept of a wait time for additional digits is
> meaningless. Progressive digit gathering implies that the signalling is
> occuring 'in-band' as would be the case with DTMF signalling on
analog
> lines.
This is incorrect. Isdn (and thus pri:s as well) can and often do use 
overlap dialing. This is very much out of band in the d channel. I have 
not seen a pri that listened inband but that may be possible as well.

Asterisk can be configured to allow overlap diling on a pri span. It will
still support enbloc diling if overlap is enabled. 
> You need to look in the Ascom and find the configuration table that lays
out
> the dialplan for the network. The PBX itself will use that information to
> determine that for example, if a number begins with '4' then it is
> 'complete' after a total of 3 digits (eg. 411) or if it begins with
1, then
> it's complete after 10 digits (1XXXNNNXXXX). After the number is
considered
> 'complete' it will build the setup frame and transmit it to the
terminating
> system. 
Fixed numbering plans are broken in so far as you cannot know the 
numbering plans of all the countries in the world. Some countries have 
variable length numbers as well. 

Use overlap dialing and let the PSTN sort things out via ss7. Asterisk 
already supports it.

Peter

I guess this topic is now closed.   The problem was that Asterisk 0.7.2 (yeah 
yeah, this is a busy production system) has buggy handling of overlapping 
extensions  ( 224  versus 2246 ).   Asterisk 1.0.3  allows such numbers to 
coexist and -does- wait to see if the dialing is completed.

In answering Peter's question, Germany definitely has a variable-length 
dialplan.   Here at our Hamburg office, the DID main number is
040 - XXX - 2040   but the DID extensions are 040 - XXX - 204XX

I agree that keeping numbers distinct is a good idea, but this box is part of a 
multi-site integration, each site has its own weird long-existing dialplan, and 
I've tried to minimise any renumbering or forcing 
people to remember prefix-numbers for special dialing.



On Tue, 7 Dec 2004, Peter Svensson wrote:
> On Tue, 7 Dec 2004, Jerry Glomph Black wrote:
>
>> jumping instantly on the first match, even tthough there is a pattern
>> _2XXX in the same dialplan, with higher priority (the 224 is in an
>> include=> context to suppress it).
>
> There is something going wrong with the ast_canmatchmore function then.
> Can you post your dialplan?
>
> A few thoughts:
>  * try putting them both in the same context.
>  * try making both either non-pattern or pattern.
>  * is there any more output from asterisk with verbosity set way up?
>
> Generally it is best to avoid shared prefixes in the dialplan, if for no
> other reason than to avoid confusion.
>