[asterisk-dev] [Code Review] more elaborate version of r1658 (not calling quit_handler/atexit twice)

Terry Wilson reviewboard at asterisk.org
Tue Jan 10 16:28:11 CST 2012


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/1662/#review5154
-----------------------------------------------------------

Ship it!


It's more moving stuff around than I normally like to do for this kind of thing in a release branch, but it fixes the issue, makes things more readable, and I haven't been able to break it. One thing to note is that previous versions would let you run "core stop when convenient" after running "core stop gracefully" as though it would work, but it really didn't because the channel.c stuff didn't get reset. So this makes the behavior cleaner even in that case.

- Terry


On Jan. 10, 2012, 3:16 p.m., wdoekes wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/1662/
> -----------------------------------------------------------
> 
> (Updated Jan. 10, 2012, 3:16 p.m.)
> 
> 
> Review request for Asterisk Developers and Terry Wilson.
> 
> 
> Summary
> -------
> 
> A more elaborate attempt at fixing issues with multiple calls to quit_handler; see https://reviewboard.asterisk.org/r/1658/
> 
> With this patch, niceness is allowed to be decreased: if you did a 'core stop gracefully' you can decrease the niceness with a 'core stop now', but not the other way around. 'core abort shutdown' is still allowed *until* the safe-shutdown part of the code has been reached.
> 
> It kind of grew larger -- and therefore less pretty -- than I had hoped unfortunately.
> 
> 
> Diffs
> -----
> 
>   branches/1.8/main/asterisk.c 350307 
> 
> Diff: https://reviewboard.asterisk.org/r/1662/diff
> 
> 
> Testing
> -------
> 
> Added a bit of sleep can_safely_quit so I could test it. Seemed to work ok.
> 
> 
> Thanks,
> 
> wdoekes
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20120110/1c20db35/attachment.htm>


More information about the asterisk-dev mailing list