[asterisk-dev] [Code Review] Fix authentication regression in res_jabber in component mode brought on by r333265.

[irroot]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/1464/#review4453
-----------------------------------------------------------



/branches/1.8/res/res_jabber.c
<https://reviewboard.asterisk.org/r/1464/#comment8597>

    This needs to be tidied up to conform with guidelines.
    
    if (something {
    } else {
    }



/branches/1.8/res/res_jabber.c
<https://reviewboard.asterisk.org/r/1464/#comment8598>

    Please while you in the code clean up white space AKA red blocks.


- irroot


On Sept. 28, 2011, 3:39 p.m., jrose wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/1464/
> -----------------------------------------------------------
> 
> (Updated Sept. 28, 2011, 3:39 p.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Summary
> -------
> 
> Component mode makes Asterisk attempt to use send_raw every time the connection times out in order to keep the jabber connection alive.  This was being skipped over by the mentioned r333265 patch in order to avoid segfaults that would happen if we tried to send messages to disconnected jabber servers.
> 
> Fixing this was primarily accomplished by splitting send_raw with a helper method so that the conditional evaluation added by r333265 can be bypassed explicitly.  In this case, since we know that we are sending a keep alive to a server we haven't already disconnected from, using aji_send_raw without the extra conditions seems safe.
> 
> 
> This addresses bug ASTERISK-18626.
>     https://issues.asterisk.org/jira/browse/ASTERISK-18626
> 
> 
> Diffs
> -----
> 
>   /branches/1.8/res/res_jabber.c 337971 
> 
> Diff: https://reviewboard.asterisk.org/r/1464/diff
> 
> 
> Testing
> -------
> 
> I tested this locally on my machine using Openfire.  I'm under the impression Michael Young did some similar testing.  In addition, Robert Dailey (the reporter) has been testing this as well and said it was working so far.  He mentioned that he'd report if anything happened that needs addressed, and it's been going for at least most of today so far.
> 
> 
> Thanks,
> 
> jrose
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20110929/a0bcd5fa/attachment.htm>