[asterisk-dev] [Code Review] 4579: res_pjsip_messaging: Serialize outgoing MESSAGEs

Matt Jordan reviewboard at asterisk.org
Fri Apr 3 11:47:33 CDT 2015


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/4579/#review15045
-----------------------------------------------------------

Ship it!


I'll just say that I enjoy that this is all it took to change the threading model in the module.

- Matt Jordan


On April 3, 2015, 11:38 a.m., Mark Michelson wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/4579/
> -----------------------------------------------------------
> 
> (Updated April 3, 2015, 11:38 a.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Bugs: ASTERISK-24937
>     https://issues.asterisk.org/jira/browse/ASTERISK-24937
> 
> 
> Repository: Asterisk
> 
> 
> Description
> -------
> 
> Because res_pjsip_messaging throws all messages to send into the threadpool, there is no guarantee that consecutive outgoing messages from the same dialplan execution will be sent in the same order.
> So for instance, if you had the following dialplan:
> 
> exten => hello,1,NoOp()
> same => n,SendMessage(hello)
> same => n,SendMessage(world)
> 
> It would be expected that the "hello" message would be sent before the "world" message. However, it cannot be guaranteed this will happen with the current threadpool usage.
> 
> The patch on this review introduces a serializer for outgoing MESSAGE requests from Asterisk. This ensures that all MESSAGE requests are sent in the same order that they are processed in the dialplan.
> 
> 
> Diffs
> -----
> 
>   /branches/13/res/res_pjsip_messaging.c 433838 
> 
> Diff: https://reviewboard.asterisk.org/r/4579/diff/
> 
> 
> Testing
> -------
> 
> The bug itself is incredibly difficult to have happen under normal circumstances, but I have confirmed that this patch has not hindered operations any.
> 
> 
> Thanks,
> 
> Mark Michelson
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20150403/182bfe3f/attachment.html>


More information about the asterisk-dev mailing list