Re: [exim] Forcing a message to be deferred if return_path h…

Top Page
Delete this message
Reply to this message
Author: Gary Palmer
Date:  
To: exim-users
Subject: Re: [exim] Forcing a message to be deferred if return_path has a problem
Tony Finch wrote:

>On Tue, 31 Jan 2006, Gary Palmer wrote:
>
>
>>I have a "custom" lookup implimented in perl which is used to set the
>>return_path variable on the outbound transports. I am wondering what is the
>>best way to force the message to be deferred if the lookup encounters a
>>problem? Returning a forced failure just means the message is delivered with
>>the old return_path, which is undesirable as it is likely unroutable outside
>>of the local system. The docs say "if it fails for another reason, delivery
>>is deferred". What would "another reason" be in this situation?
>>
>>
>
>A non-forced string expansion failure. From the perl point of view,
>returning undef causes a forced failure whereas calling die (I think) is
>an unforced failure.
>
>


Interesting. Thanks for the hint Tony, that does indeed have the
desired effect. Never thought to try "die" before, since that tends to
kill the process. I guess embeded perl traps that somehow.

Thanks again

Gary