How soon will this make it into an update? And what module should I look for the update in? Core? Which version?
it’s in core and checked into 2.9 and 2.10 but I don’t plan on publishing it until we get some testing. You can grab it from SVN or just patch your existing functions.inc.php.
However, it’s not clear from the comments that gherbstman made whether we are always triggering the issue or whether it really is something that needs to be configurable in Asterisk.
However, it would be great to get some testing on a system that has this issue and confirmation if this helps or not. (And it would be useful if this is related to have a ticket open to that affect and we can reference this changeset in the ticket.
Philippe,
I’d LOVE to test it.
Can you tell me how to grab it from SVN or to patch my existing functions.inc.php?
Philippe,
As far as testing goes, you may e-mail or call me and I’ll be happy to do whatever you want me to do in order for your track down the issue. I’ll send you a private e-mail right now with my contact information.
You can also easily duplicate the issue yourself. Just take a basic FreePBX Distro install and configure a Callcentric Trunk (I have a “spare” account if you don’t have one). Use these trunk parameters:
http://www.callcentric.com/support/device/freepbx
In addition to these parameters, you’ll want to include
sendrpid=yes
in your trunk settings, which Callcentric requires if you want to send your own caller ID.
Configure the inbound route to go to an announcement, and the on to an extension.
When you call in, the announcement will play, and then the extension will ring. However, when you pick-up the extension, the call will terminate.
I have a logfile with verbosity at 10 and debug on that covers two minutes. During that two minutes, a call comes in and goes through this exact process with nothing else going on.
My only hesitation in putting it on a bug report is that it may contain passwords, phone numbers, or other private information, but I’d be happy to e-mail it to you directly.
Note: I’ve edited the comment above, so if you were notified of the original by e-mail, be sure to read the latest version on the forum before you begin testing.
I’ve also opened a bug report per your request.
This problem, like so many I’ve experienced lately, seems to have resolved itself with no intervention on my end. It may be that Callcentric modified its code to ignore the reinvites, but I’m beginning to suspect that there was some kind of backbone problem that was remedied, and that CC’s diagnosis was wrong.
On a possibly related note, I was having a ton of call routing problems (numbers that were getting circuits busy or disconnect recordings but were known to be good numbers and which worked on other provides) over the last few days as well.
We maintain several TDM overflow routes and I can tell you this week has been of the worst performing weeks for at least 1/2 dozen major wholesale core providers.
The FCC changes the way ILEC’s can pay reciprocal compensation in rural/independent rate centers and the carriers are scrambling to change billing. I have found that most carriers are simply sending 503’s on routes they don’t want to accept traffic.
This issue seems to have returned. Now, if I have sendrpid=yes on Callcentric, whenever I place a call on hold, the call drops.
On the plus side, most of my call routing issues seem to be resolved…