[exim-dev] [Bug 266] Accept mail after local_scan() crash in…

Pàgina inicial
Delete this message
Reply to this message
Autor: Phil Pennock
Data:  
A: exim-dev
Assumpte: [exim-dev] [Bug 266] Accept mail after local_scan() crash instead of defer
------- You are receiving this mail because: -------
You are the QA contact for the bug.

http://bugs.exim.org/show_bug.cgi?id=266

Nigel Metheringham <nigel@???> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|Exim 4.72                   |Exim 4.73


Phil Pennock <pdp@???> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pdp@???
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX





--- Comment #3 from Phil Pennock <pdp@???> 2010-06-06 06:38:10 ---
Fundamentally bad idea. If local_scan is crashing, we can't guarantee that we
have the data unmolested and can't meet the requirement to not arbitrarily lose
data. The *only* safe way is to cause the last place which has the data
known-intact to retry later.

If someone has flaky code, it's probably best to package it up as a daemon
which can be talked to remotely via ${readsocket} or ${run} or av_scanner
cmdline and handle things with some degree of isolation. But for code which
crashes inside the Exim process space, this is a definite WONTFIX.


--
Configure bugmail: http://bugs.exim.org/userprefs.cgi?tab=email