Summary: | ASTERISK-03280: [Patch] Loss of callerid when calling from chan_vpb.c | ||
Reporter: | tcord (tcord) | Labels: | |
Date Opened: | 2005-01-15 04:02:44.000-0600 | Date Closed: | 2005-01-17 21:43:16.000-0600 |
Priority: | Minor | Regression? | No |
Status: | Closed/Complete | Components: | Core/General |
Versions: | Frequency of Occurrence | ||
Related Issues: | |||
Environment: | Attachments: | ( 0) patch ( 1) patch ( 2) patch ( 3) patch ( 4) patch | |
Description: | Callerid is not maintained when calling from vpb channels ****** ADDITIONAL INFORMATION ****** patch to fix problem included | ||
Comments: | By: nick (nick) 2005-01-15 10:06:28.000-0600 Do you have a disclaimer? By: Mark Spencer (markster) 2005-01-15 12:24:57.000-0600 We'll need the disclaimer as well as a patch for CVS head, thanks! By: tcord (tcord) 2005-01-16 01:48:50.000-0600 Disclaimer has been emailed as PDF to malcolmd@digium.com, don't have a fax near me right now (snail mail will take a month from here) By: Mark Spencer (markster) 2005-01-16 01:56:07.000-0600 Perfect, do you have a patch for CVS head, too? By: tcord (tcord) 2005-01-16 05:23:26.000-0600 patch for CVS head attached By: tcord (tcord) 2005-01-16 06:25:07.000-0600 woops - better hold that patch it is for v1.0.12, will need some changes for current CVS By: tcord (tcord) 2005-01-16 08:14:00.000-0600 Final patch gives the ability to represent callerid as "String"<number> consistent with other drivers By: VoiceTronix (benkramer) 2005-01-16 19:06:08.000-0600 Have you tested that this was an issue with the CVS head version of Asterisk before producing the patch ? The reason I ask this is that the variable me->callerid is not guaranteed to have anything in it, as it is no longer used to store the CID, but may have temporary information in it. By: tcord (tcord) 2005-01-16 20:43:30.000-0600 me->callerid is obtained from struct vpb_pvt and in this instance comed from the config file vpb.conf lines callerid="...", if it contains nothing the code as presented still works, it just inserts "unknowm" Do you mean I should perform a preliminary check as I have presented in the last patch (which does nothing?) so ignore it please. This patch is purely to fix problems in outbound calls from chan_vpb devices. The "callerid" lines from vpb.conf go nowhere unless they are copied into the ast_channel structure somewhere during a call, do you have any better suggestions where this should be done or have I missed the point of your comment? The patch as represented 01-16-05 08:09 has been tested and fixes the issue outlined above By: VoiceTronix (benkramer) 2005-01-16 21:10:40.000-0600 There is actually a confusion as to the current purpose and original purpose of the "callerid" config option in vpb.conf. I will look into clarifying this soon. So you patch provides which one: A) the "Station identification" when calling from a vpb station port to another resource ? B) forwards the "caller id" that was collected on when the call was presented to the system ? This will help me to integrate your patch and clarify the usage of the "callerid" config option. By: tcord (tcord) 2005-01-16 21:57:35.000-0600 You are correct with (A), there was no facility in chan_vpb to present a callerid when calling from a vpb channel to another resource, I haven't tested the (B) option yet but will get onto this one soon. I will only be able to test the Australian stuff, but a quick look through the code indicates this is all in place anyway callerid=on|bell|v23 is what you put on FXO lines callerid="Name"<number> on FXS lines The callerid config option as far as I can see is not documented, but if this info is defined prior to channel=N in the config file it works with this patch. By: VoiceTronix (benkramer) 2005-01-16 22:23:33.000-0600 can you contact me directly at ben at voicetronix.com.au By: VoiceTronix (benkramer) 2005-01-17 21:27:39.000-0600 I have patched this and cleaned up the caller ID stuff a bit. It should also forward caller ID information from a trunk if the trunk is not answered till after the CID collection has completed (ie ~ do a wait(3) before Dial()ing). Tested with OpenLine4 and Grandstream BudgeTone100. By: VoiceTronix (benkramer) 2005-01-17 21:29:20.000-0600 See last comment. |