[asterisk-dev] [Code Review] 2878: app_queue: Make manager events tolerant of Local channel shenanigans

[Mark Michelson]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/2878/#review9831
-----------------------------------------------------------

Ship it!


Ship It!

- Mark Michelson


On Sept. 24, 2013, 9:49 p.m., Matt Jordan wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/2878/
> -----------------------------------------------------------
> 
> (Updated Sept. 24, 2013, 9:49 p.m.)
> 
> 
> Review request for Asterisk Developers and Mark Michelson.
> 
> 
> Bugs: ASTERISK-22507
>     https://issues.asterisk.org/jira/browse/ASTERISK-22507
> 
> 
> Repository: Asterisk
> 
> 
> Description
> -------
> 
> app_queue does an admirable job of attempting to manage the Lovecraftian nightmare that are Local channel optimizations. Sometimes, however, even it can't defeat the horror.
> 
> Consider the following scenario:
>  SIP/foo <-> L;1 <-> L;2 <-> SIP/agent
> 
> SIP/agent answers, triggering a Local channel optimization. This is fine - app_queue will normally do the following:
>  * Listen for the Local optimization events and update our agent accordingly to SIP/agent
>  * When we get a hangup, publish the AgentComplete event based on our information (SIP/foo and SIP/agent)
> 
> However, as with all things that depend on sanity from something as capricious as Local channels, things can go wrong:
>  (1) SIP/agent immediately hangs up upon answering, because "customer service is our #1 priority" doesn't apply on casual Fridays
>  (2) Asterisk is either (a) Running under valgrind; (b) Running under gdb; (c) Running on Richard's dev machine/some other ancient piece of equipment that should have been put out of its misery ages ago
> 
> In that case, the messages *may* arrive to app_queue in the following order:
>  * Hangup SIP/Agent
>  * Hangup SIP/foo
>  * Optimize L;1/L;2
>  * Hangup L;2
>  * Hangup L;1
> 
> When we receive the hangup of the agent or the caller, we'll attempt to publish the AgentComplete event. However, we now have a problem - we think our agent is L;1. We haven't received the optimization messages - but L;1 is already dead, so any attempts to query it from the Stasis cache result in a big fat NULL.
> 
> There are two options here:
>  (1) If we can't retrieve channel information, bail
>  (2) If we can't retrieve channel information, publish what we know
> 
> This patch goes with option #2. We still have a lot of pertinent queue information (interface, queue name, etc.) - and we may as well tell the user what we know. They will still at least get the "AgentComplete" event, which "completes" the known Agent information.
> 
> 
> Diffs
> -----
> 
>   ./branches/12/apps/app_queue.c 399746 
> 
> Diff: https://reviewboard.asterisk.org/r/2878/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Matt Jordan
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20130927/7cdcba95/attachment-0002.html>