asterisk users - Jun 2003 - T100P + Capi + Caller ID issue

Hi 2 all.

I've a nice issue my * system.
I have:
asterisk latest cvs
T100P + Zhone cb (3fxo+10fxs used)
Diva Server 4BRI running Capi
a bunch of pots and sip phones

The whole system is running good (thanks to capigod
and Bic. for the help with the Eicon Diva), but there's
a very strange issue when calls arrive on the Diva
and then bridged to the pots on the channel bank:
the voice is not perfect on my side (I can hear some pops on
the phone side), but is perfect on the Capi side.
If I answer on a sip phone, all is ok, good voice , both dirs.

But then those "pops" seems to break Caller ID : callerID
is received by asterisk on the capi side (I can debug it,
and run an agi script for it),but transmitted bad to the pots.
What I see on my analog phone is a messed up CID (strange chars or
number if any). If I answer with a sip phone, all is ok
(voice good, CID is ok on the sip side and so on).

I thought that could be a issue with the zhone or the T100P.
I have changed either the zhone and the T100P (since I have
2 of them), but the issue remains.

But, if a call is received not via Capi, but via an analog line
on the channel bank, the caller id is display correctly!
(the zhone doesn't pass CID, but my agi script detects that
 and set cid to "UNKNOWN" <0> ).

Of course, CID between local extension (through the zhone)
is ok....

So I've done a test : I added a TDM400B to my setup and
connected 4 pots to it... et voil?, CID now works ok
on these phones (even if the call is coming from the CAPI side).

So any idea on that???

All the cards doesn't share any irq, the system is a debian 3.0r1
with kernel 2.4.20 , P4 2.4 Ghz and 512 Mb Ram.

seems that there's some skew between the capi side and the T100P
side.. but is just an idea....

matteo.







-- 
Matteo Brancaleoni
Powered by RedHat Linux 8.0
Linux User #153521
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GS d? s:- a- C+++ UL++++ P+ L+++ E- W+++ N++ o K- w--
O- M-- V-- PS PE- Y PGP++ t 5 X+ R tv- b++ DI D+
G e h! r++ y
------END GEEK CODE BLOCK------

I've posted 2 shots of the CID spectrum on
http://vmail.espia.it/shots/ , for those
that want to dig more in the issue.

Matteo.

Il gio, 2003-06-05 alle 16:12, Brancaleoni Matteo ha
scritto:
> Hi 2 all.
> 
> I've a nice issue my * system.
> I have:
> asterisk latest cvs
> T100P + Zhone cb (3fxo+10fxs used)
> Diva Server 4BRI running Capi
> a bunch of pots and sip phones
> 
> The whole system is running good (thanks to capigod
> and Bic. for the help with the Eicon Diva), but there's
> a very strange issue when calls arrive on the Diva
> and then bridged to the pots on the channel bank:
> the voice is not perfect on my side (I can hear some pops on
> the phone side), but is perfect on the Capi side.
> If I answer on a sip phone, all is ok, good voice , both dirs.
> 
> But then those "pops" seems to break Caller ID : callerID
> is received by asterisk on the capi side (I can debug it,
> and run an agi script for it),but transmitted bad to the pots.
> What I see on my analog phone is a messed up CID (strange chars or
> number if any). If I answer with a sip phone, all is ok
> (voice good, CID is ok on the sip side and so on).
> 
> I thought that could be a issue with the zhone or the T100P.
> I have changed either the zhone and the T100P (since I have
> 2 of them), but the issue remains.
> 
> But, if a call is received not via Capi, but via an analog line
> on the channel bank, the caller id is display correctly!
> (the zhone doesn't pass CID, but my agi script detects that
>  and set cid to "UNKNOWN" <0> ).
> 
> Of course, CID between local extension (through the zhone)
> is ok....
> 
> So I've done a test : I added a TDM400B to my setup and
> connected 4 pots to it... et voil?, CID now works ok
> on these phones (even if the call is coming from the CAPI side).
> 
> So any idea on that???
> 
> All the cards doesn't share any irq, the system is a debian 3.0r1
> with kernel 2.4.20 , P4 2.4 Ghz and 512 Mb Ram.
> 
> seems that there's some skew between the capi side and the T100P
> side.. but is just an idea....
> 
> matteo.
> 
> 
> 
>