[asterisk-bugs] [JIRA] (ASTERISK-21147) asterisk doesn't continue to the next priority after a soft hangup

[Rusty Newton (JIRA)]

Rusty Newton created ASTERISK-21147:
---------------------------------------

             Summary: asterisk doesn't continue to the next priority after a soft hangup
                 Key: ASTERISK-21147
                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-21147
             Project: Asterisk
          Issue Type: Bug
      Security Level: None
          Components: PBX/General
    Affects Versions: 1.8.20.1, 10.12.1, 11.2.1
         Environment: Linux manage 3.0.6-gentoo #1 SMP Wed Nov 16 02:09:24 Local time zone must be set--see zic  x86_64 Intel Core i7 9xx (Nehalem Class Core i7) GenuineIntel GNU/Linux
            Reporter: Paul Albrecht


Noticed a change in ast_softhangup after moving to asterisk 10. Asterisk doesn't continue to the next priority after a soft hangup which doesn't make much sense.

See asterisk-dev: http://lists.digium.com/pipermail/asterisk-dev/2013-February/058709.html

[Edit by Rusty Newton - adding the E-mail below (links could get broken in the future)]

{quote}
> >> I noticed a change in ast_softhangup after moving to asterisk 10.
> >> Now
> >> asterisk doesn't continue to the next priority after a soft hangup
> >> which doesn't make much sense. Is this an feature or bug?
> > 
> > Either svn revisions
> > -r190423 Changes channels to ref counted objects
> > -r225244 Add support for OBJ_MULTIPLE
> > 
> > Causes the issue.  The app does not continue in the dialplan if no
> > channels are found to hangup or there was a memory allocation
> > error.
> > 
> > Its a bug.
> 
> That's what I thought. What about ast_channel_clear_softhangup? Why
> does it need to go through the readq?
>  The only thing ast_softhangup queues is a null frame and I thought
>  they're supposed to be ignored.

There is a comment in ast_channel_clear_softhangup() that says why.
There is another comment in __ast_read() where the AST_CONTROL_END_OF_Q
control frame is put in the read queue that gives even more information.

Basically, when ast_read() sees a hangup, there may be pending frames
in the read queue so it puts a control frame in the read queue to know
when all pending frames have been read and to block any more frames
from being put on the read queue.  ast_read() will then return a NULL
to indicate a hangup.
{quote}


--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira