[asterisk-users] chan_sip deadlocks after some time

2008-01-22 Thread Henning Holtschneider
 if in prospect of success.

Thanks for your help,
Henning Holtschneider
--
LocaNet oHG - http://www.loca.net
Lindemannstrasse 81, D-44137 Dortmund
tel +49 231 91596-25, fax +49 231 91596-55
sip [EMAIL PROTECTED]

Registergericht Amtsgericht Dortmund HRA 14208
Geschäftsführer Sven Haufe, Henning Holtschneider


signature.asc
Description: PGP signature
___
-- Bandwidth and Colocation Provided by http://www.api-digital.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] DECT to SIP gateway experiences

2007-03-15 Thread Henning Holtschneider
Am Thu, 15 Mar 2007 14:32:28 +1100
schrieb Daniel Pittman [EMAIL PROTECTED]:

 It seems to me that a direct SIP to DECT gateway could have
 significant advantages in terms of supporting the MWI (voicemail)
 indicator on the DECT phone directly -- there just isn't any way I
 could trigger it on any of the analog sets I have at the moment.
 
 Unfortunately I can't local any information on this; the documentation
 for the Zyxel DECT gateways and Siemens Gigaset DECT bases don't say
 *anything* about their supporting MWI hardware from a SIP server.

The MWI depends on the vendor who implements this feature. IIRC, MWI is
not part of the DECT specifiations so you won't find any generic
information regarding this topic.

 The other killer feature that a DECT base could theoretically offer is
 some sort of soft menu system -- ADSI, XML, or whatever.

Theoretically! The soft menus that you see on DECT handsets are
proprietary so they don't work with 3rd party DECT base stations.

 So, can anyone comment on support for MWI in the SIP DECT gateways?
 How about soft menu support?

MWI works on the KIRK Wireless gateways we are using. There is support
for soft menus which come from the DECT base but they cannot be sent by
the SIP server. MWI only works on KIRK handsets; our attempts with
Siemens and Philips handsets were unsuccessful.

Best regards,
Henning Holtschneider
--
LocaNet oHG - http://www.loca.net
Lindemannstrasse 81, D-44137 Dortmund
tel +49 231 91596-25, fax +49 231 91596-55
sip [EMAIL PROTECTED]

Registergericht Amtsgericht Dortmund HRA 14208
Geschäftsführer Sven Haufe, Henning Holtschneider


signature.asc
Description: PGP signature
___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] T.38 faxing with spandsp and Grandstream HT.486

2006-11-01 Thread Henning Holtschneider
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Johann Steinwendtner schrieb:

 I 'm trying T.38 faxig with spandsp using rxfax/txfax as fax terminal.
 As another endpoint I 'm using Grandstream HT 486 ATA and a fax machine.
 Has anybody success with the HT486 as T.38 terminal ?

I've successfully faxed via T.38 using these combinations:

fax machine - HT486 - SIP server - HT486 - fax machine

fax machine - HT486 - SIP server - Thomson ST620 - fax machine

fax machine - Thomson ST620 - SIP server - HT486 - fax machine

fax machine - AVM Fritzbox 7050 - SIP server - HT486 - fax machine

fax machine - AVM Fritzbox 7050 - SIP server - Thomson ST620 - fax
machine

(note: T.38 is not officially supported on the AVM Fritzbox. It only
works with older firmware versions and only when sending. When receiving
a fax tone from the remote end, the Fritzbox does not issue a T.38
re-invite).

I also just recently tested the following cases:

fax machine - HT486 - OpenPBX with spandsp T.38 - RxFAX

fax machine - Thomson ST620 - OpenPBX - RxFAX

TxFAX - OpenPBX - HT486 - fax machine

TxFAX - OpenPBX - Thomson ST620 - fax machine

 ATA as originator: I managed only onetimes a successfull T.38 fax
 session. The other times the HT486 did not initiate a re-invite with
 T.38 parameters. Or shall the Terminator inititate a re-invite ?

When I tested T.38 with OpenPBX, it never detected the fax tone from the
 fax machine and thus did not issue a T.38 re-invite. This might be due
to a configuration problem because I'm not overly familiar with OpenPBX
and there are not many user reports available on the net. If you point
me to the necessary patches for Asterisk, I will be glad to repeat my
tests ;-)

The fax tone detection and T.38 re-invite is always performed by the
receiving terminator or gateway. I haven't read the corresponding RFC
thoroughly enough but this is the way it works on all ATAs and gateways
I've worked with so far. I also noticed that many modems which are
connected to or built into PCs do not send a proper fax tone (CNG) so
make sure you are using a good old fax machine. Thirdly, make sure that
your fax machine does not support V.34 (class 2.1 aka Super G3)
because T.38 is limited to 14400 bps on the modem side. I only tested
one V.34 fax machine but its modem would not negotiate with the HT486
properly.

 txfax as originator: T.38 fax exchange takes place but the transmission
 is not successful, txfax reports errorcode 60 (Disconnected after
 permitted retry).

I haven't been able to send more than a simple page using TxFAX with
T.38 support yet. And even that did not work reliably. Either, the
single page wasn't transmitted at all or fax machine on the HT486/ST620
recognized the end of the page but failed to negotiate the end of the
transmission correctly.

Once again, I only tried OpenPBX with T.38 termination support, yet, so
your problem may be different.

 Can someone recommend a T.38 able ATA which is working with spandsp ?

T.38 support in spandsp is still work-in-progress so I think it's a
litte early to  make any recommendations. All I can say is that the T.38
support on the HT486 works reliably if the remote end does not stretch
the specifications too much.

Cheers,
Henning Holtschneider
- --
LocaNet oHG - http://www.loca.net
Lindemannstrasse 81, D-44137 Dortmund
tel +49 231 91596-25, fax +49 231 91596-55
sip [EMAIL PROTECTED]
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (MingW32)

iD8DBQFFSLloP9goCV2uudcRAg6dAJ9QQrTBvyCt7vEPO4YV+kXvfHsn6wCgv/5I
EKunq1uf2+mI0+pjm0+yeAw=
=TFn1
-END PGP SIGNATURE-

___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] * Meetme Freeze patch found

2006-03-31 Thread Henning Holtschneider
On Monday 27 March 2006 20:17, Benoit Panizzon wrote:
 On Friday 24 March 2006 16:05, Benoit Panizzon wrote:
  Hi all
 
  Apparently there is a patch for those 1.2.4/5 MeetMe Freezes:
 
  http://bugs.digium.com/view.php?id=5884
 
  Haven't tried it out yet.

 I can now confirm: No freezes/crashes anymore since I applied the patch.

It is still possible to crash the Asterisk machine, at least with Asterisk 
1.2.4 and the patch applied: 

1. use iLBC codec
2. dial into conference
3. while Allison talks to you, put line on hold
4. repeat steps 2 and 3 several times

On our production machine, Allision's voice starts to be choppy on the fifth 
channel and the machine freezes after answering the sixth channel.

I had no time to analyze this more thoroughly, but disallowing iLBC fixed the 
problem, so it might also be an iLBC codec problem.

Cheers,
Henning Holtschneider
--
LocaNet oHG - http://www.loca.net
Lindemannstrasse 81, D-44137 Dortmund
tel +49 231 91596-25, fax +49 231 91596-55


pgpQ7t05UDNNm.pgp
Description: PGP signature
___
--Bandwidth and Colocation provided by Easynews.com --

Asterisk-Users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] MeetMe freezes machine with Junghanns QuadBRI cards

2006-03-24 Thread Henning Holtschneider
On Thursday 23 March 2006 21:14, stoffell wrote:
 On 3/23/06, Henning Holtschneider [EMAIL PROTECTED] wrote:
  I've got Asterisk 1.2.4 running with two Junghanns QuadBRI cards using
  the qozap driver from bristuff 0.3.0-PRE-1l. One of the cards is running
  in TE mode, the other one in NT mode.

 Have you (or can you) tried it with 0.3.0-pre1k ?

Yes, the problem occurs with 0.3.0-pre1k, too. I will follow BJ Weschke's 
advice and try the latest SVN tonight. I will report success or failure to 
the list.

Cheers,
Henning Holtschneider
--
LocaNet oHG - http://www.loca.net
Lindemannstrasse 81, D-44137 Dortmund
tel +49 231 91596-25, fax +49 231 91596-55


pgpL0Yx6AmCIT.pgp
Description: PGP signature
___
--Bandwidth and Colocation provided by Easynews.com --

Asterisk-Users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] RE: MeetMe freezes machine with Junghanns

2006-03-24 Thread Henning Holtschneider
On Thursday 23 March 2006 22:14, BJ Weschke wrote:

  There's been two very recent commits (one less than an hour ago) that
 may very well correct your issues.

The patch at http://bugs.digium.com/view.php?id=5884 fixes the problem!

Cheers,
Henning Holtschneider
--
LocaNet oHG - http://www.loca.net
Lindemannstrasse 81, D-44137 Dortmund
tel +49 231 91596-25, fax +49 231 91596-55


pgpoDln7bEb6t.pgp
Description: PGP signature
___
--Bandwidth and Colocation provided by Easynews.com --

Asterisk-Users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


[Asterisk-Users] MeetMe freezes machine with Junghanns QuadBRI cards

2006-03-23 Thread Henning Holtschneider
Hello everybody,

I've got Asterisk 1.2.4 running with two Junghanns QuadBRI cards using the 
qozap driver from bristuff 0.3.0-PRE-1l. One of the cards is running in TE 
mode, the other one in NT mode.

Whenever I call into an empty MeetMe conference room on one of the NT ports or 
via SIP and hang up the call during the you are currently the only 
participant in this conference greeting, the Asterisk server freezes. I can 
still ping the machine, but all userspace processes, system logging, kernel 
logging etc. are hung. The machine takes about 15 minutes to recover from 
this state. At that point, the system load has risen to about 15-20.

If I call into the server on one of the TE ports while the system is frozen, 
it instantly un-freezes and resumes normal operation.

Does anyone else experience the same problem?

Cheers,
Henning Holtschneider
--
LocaNet oHG - http://www.loca.net
Lindemannstrasse 81, D-44137 Dortmund
tel +49 231 91596-25, fax +49 231 91596-55


pgpx7NTEI4KJo.pgp
Description: PGP signature
___
--Bandwidth and Colocation provided by Easynews.com --

Asterisk-Users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users