On Mon, 17 Dec 2001, Greg A. Woods wrote:
> The issue of dynamically loaded code, at least in what I imagine is the
> way it might be used in this particular case, isn't a matter of
> convenience, and the risks are very much higher than you make them out
> to be.
That is something that I am going to have to judge when I get round to
looking at this particular issue. At present I am not qualified to make
a judgement.
But (*without* really knowing much about it) it seems to me that if you
are running a host on which nobody except yourself logs in (a typical
mail switch) and on which you therefore have complete control of all
processes that are run, and how they are run, the risks posed by dynamic
loading are likely to be a lot less than on a host where there are many
login accounts for random people.
In any event, *if* I add code to Exim, it would be conditional, and have
to be specially requested at compile time.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
