[asterisk-dev] [Code Review] Manager originate: Don't fall back to "s" when given extension does not exist

Olle E Johansson reviewboard at asterisk.org
Mon Sep 19 13:07:49 CDT 2011


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/1445/
-----------------------------------------------------------

(Updated Sept. 19, 2011, 1:07 p.m.)


Review request for Asterisk Developers.


Changes
-------

Review 1446 for version 10 suggests a more long-term patch for ast_pbx_start()


Summary
-------

Currently, if a manager originate tries to access an extension that does not exist, pbx.c ast_pbx_start makes the call fall over to "s at default" which should *not* be the case.

We agreed on IRC to fix pbx_start later, but this patch (aimed at 1.8) fixes the manager command so that (in the manner of the SIP channel) we check for the extension before trying the call. This will affect current behaviour and should be documented properly. It may affect applications that rely on this - users may have built dialplans that use "s" this way instead of fixing a fallback within their application. We might want to have an option in manager.conf to preserve old behaviour for this reason. 

Separately, we should check other originate (CLI, app) as well, but let's separate the patches. 

Background: extension "s" should ONLY be used when NO extension is given. It's not a wildcard and not a fallback when things go wrong. 

Side note: The bug exists in 1.4 too, so in all 1.6.x releases.


Diffs
-----

  /branches/1.8/main/manager.c 336574 

Diff: https://reviewboard.asterisk.org/r/1445/diff


Testing
-------

Feels like the code compiles on OS/X. This review is more about if we want to fix this in 1.8 and how. After that, I'll test the code.


Thanks,

Olle E

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20110919/ee75bd07/attachment-0001.htm>


More information about the asterisk-dev mailing list