[asterisk-dev] daemonising after possible errors

[Tzafrir Cohen]

On Tue, Dec 05, 2006 at 02:09:44PM +0200, Tzafrir Cohen wrote:
> Hi
> 
> Something that I find bothering when I try to debug Asterisk is that it
> deamonizes before most errors can occour. It will fork into background
> before many potential fatal errors occour. Such fatal errors are in the
> module loading time.
> 
> Let's assume for the sake of discussion now that moving the 
> daemonization time is simple.
> 
> What do you think about the following:
> 
> * Moving the daemonization to the end of the modules loading
> 
> * Making it optional to move it there. I recall that there was such an
>   option with bind or squid, but I'm not sure.

I see practically no responses to that.

Consider, for instance, http://bugs.digium.com/view.php?id=7290 .

I have suggested that the handling of some specific module errors would
not be to crash Asterisk. This is also because there is no decent
handling to those kind of crashes. So either keep working with chan_zap
mostly disabled (use zap restart), or crash, but in a way that the
sysadmin can handle it.

-- 
               Tzafrir Cohen       
icq#16849755                    jabber:tzafrir at jabber.org
+972-50-7952406           mailto:tzafrir.cohen at xorcom.com       
http://www.xorcom.com  iax:guest at local.xorcom.com/tzafrir