Re: [Exim] Problem with -Meb environment being destroyed

Top Page

Reply to this message
Author: Philip Hazel
Date:  
To: Phil Pennock
CC: Exim Users
Subject: Re: [Exim] Problem with -Meb environment being destroyed
On Sun, 21 May 2000, Phil Pennock wrote:

> Looking at the source, I see that the environment is explicitly
> destroyed in exim.c, with a comment starting:
> /* Destroy the environment. Exim makes no use of any of it
>
> *coughs* Does the documentation need updating, or do VISUAL and EDITOR
> need rescueing(sp?)?


Ooops. Damn. This is my fault. One half of brain not paying attention to
the other half.

The destroying of the environment is the only way I know - and it works
only on some operating systems - of getting Exim to write its timestamps
in "wallclock" time. I introduced it after a complaint that if the TZ
variable was set by some user who then called Exim, it logged in the
wrong timezone.

I thought "Exim doesn't use any environment variables, so it's safe". I
completely forgot about EDITOR and VISUAL. Sigh. Exim is getting too
big, or maybe my brain is just fading. Again. :-)

My immediate reaction to this is to back off the destroying of the
environment, especially since it works only for some operating systems
(in the others it forces GMT rather than wallclock time). Or maybe it
should not be the default?

You can compile Exim without it already, by putting

#define HANDS_OFF_ENVIRONMENT

in the OS/os.h-xxxx file, where xxxx is your operating system. Or just
patch the source, of course.

Another alternative, is to avoid destroying the environment specifically
in the -Meb case. (And any others which are relevant.) Maybe that makes
more sense.

Thanks for the report. I will do something about it in due course.

Philip

-- 
Philip Hazel            University of Cambridge Computing Service,
ph10@???      Cambridge, England. Phone: +44 1223 334714.