[asterisk-dev] [Code Review] 4468: Various: bugfixes discovered through use of chaos

Mark Michelson reviewboard at asterisk.org
Mon Mar 9 17:29:03 CDT 2015


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/4468/#review14618
-----------------------------------------------------------



/branches/13/main/config.c
<https://reviewboard.asterisk.org/r/4468/#comment25175>

    Handling the possibility of cloned being NULL is necessary, but the problem with just returning here is that it leaves new in a half-altered state.
    
    I think the way to do this is to perform all necessary allocations up front and then perform the necessary changes to new once they have all succeeded.


- Mark Michelson


On March 9, 2015, 8:50 p.m., Scott Griepentrog wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/4468/
> -----------------------------------------------------------
> 
> (Updated March 9, 2015, 8:50 p.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Repository: Asterisk
> 
> 
> Description
> -------
> 
> Using the CHAOS random failure mechanism (https://reviewboard.asterisk.org/r/4463/) I uncovered several possibilities of a crash on null pointer, and one uninitialized variable that allowed ast_random to lock up if used before utils init.
> 
> Additionally, extra output to determine why Asterisk failed to initialize has been added.
> 
> 
> Diffs
> -----
> 
>   /branches/13/main/xmldoc.c 432661 
>   /branches/13/main/utils.c 432661 
>   /branches/13/main/endpoints.c 432661 
>   /branches/13/main/config.c 432661 
>   /branches/13/main/codec_builtin.c 432661 
>   /branches/13/main/asterisk.c 432661 
> 
> Diff: https://reviewboard.asterisk.org/r/4468/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Scott Griepentrog
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20150309/17a22a23/attachment.html>


More information about the asterisk-dev mailing list