[asterisk-dev] [Code Review] Don't miss control frames if a call is answered and hung up very quickly

[Terry Wilson]

> On 2010-06-25 09:56:38, David Vossel wrote:
> > This fixes that weird case that I mentioned earlier, and this looks correct to me.  There is one thing that kind of bothers me that is completely unrelated to your changes.  If the _softhangup variable is a bitfield, then why do we do things like (c0->_softhangup === AST_SOFTHANGUP_UNBRIDGE) in the ast_generic_bridge function?

That is a good question. Another one is why ast_queue_hangup does a trylock before setting chan->_softhangup but, if we fail to get the lock _softhangup never gets set. Every step where I looked at this bug ended up uncovering something else really weird. This is code that "has always been there" so, yeah...


- Terry


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/740/#review2280
-----------------------------------------------------------


On 2010-06-24 15:17:47, Terry Wilson wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/740/
> -----------------------------------------------------------
> 
> (Updated 2010-06-24 15:17:47)
> 
> 
> Review request for Asterisk Developers, Russell Bryant and David Vossel.
> 
> 
> Summary
> -------
> 
> When an outgoing call is answered and hung up by the far end *very* quickly, we may not read any frames and therefor end up with a call that displays the wrong disposition/DIALSTATUS. The reason is because ast_queue_hangup() immediately sets the _softhangup flag on the channel and then queues the HANGUP control frame, but __ast_read refuses to read any frames if ast_check_hangup() indicates that a hangup request has been made (which it will if _softhangup is set). So, we end up losing control frames.
> 
> This change makes __ast_read continue to read frames even if a soft hangup has been requested. I believe this should be safe as we actually check farther down in __ast_read for HANGUP frames and return NULL just like we would have if we skipped the read via the goto done statement, and we don't call into any of the channel tech callbacks unless the readq is empty.
> 
> 
> This addresses bug 16946.
>     https://issues.asterisk.org/view.php?id=16946
> 
> 
> Diffs
> -----
> 
>   /trunk/include/asterisk/channel.h 272414 
>   /trunk/main/channel.c 272414 
> 
> Diff: https://reviewboard.asterisk.org/r/740/diff
> 
> 
> Testing
> -------
> 
> I have verified via sipp that when placing an outbound call that is answered and immediately hung up that the call shows up as answered (where it doesn't w/o the patch). I have also run the testsuite and verified that all tests continue to pass.
> 
> 
> Thanks,
> 
> Terry
> 
>