Jeff Rife wrote: > So, is the multiplexor now creating a Perl process for each milter > callback, or for each e-mail or what?
The performance hit for embedded perl is not that great, actually. CPU performance should be almost the same during scanning. Startup costs will be more expensive, because you're forking and execing a Perl interpreter and reinitializing SpamAssassin, but that only happens every 100 requests (or whatever you've set MX_REQUESTS to). You also use more memory, but again, the overhead is not terrible. It turns out that Perl is very *poor* at sharing memory among child processes post-fork. It uses a reference-counted garbage collector, so even looking at a piece of data probably modifies the reference count, dirtying the page and eliminating the possibility of sharing. :-( Regards, David. _______________________________________________ NOTE: If there is a disclaimer or other legal boilerplate in the above message, it is NULL AND VOID. You may ignore it. Visit http://www.mimedefang.org and http://www.roaringpenguin.com MIMEDefang mailing list [email protected] http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
