> Date: Fri, 24 Sep 2004 12:45:16 -0400 (EDT) > From: "David F. Skoll" <[EMAIL PROTECTED]> > Subject: Re: [Mimedefang] Reject vs Drop and MX > To: [EMAIL PROTECTED] > Message-ID: > <[EMAIL PROTECTED]>presence > Content-Type: TEXT/PLAIN; charset=US-ASCII > > On Fri, 24 Sep 2004, Troy Carpenter wrote: > >> According to that section, the first entry says that if I define my >> global >> outside of all routines, it is available to all routines. Does that >> imply >> that if I change that global variable within a routine that my change >> doesn't keep after the routine ends? > > It keeps in the original slave, but the problem is that the > filter_begin might happen in a completely different Perl slave from > the corresponding filter_relay. On a quiet mail server, this happens > occasionally. On a busy one, it happens all the time.
Ah, that explains that...my system is VERY low volume. I normally have only two slaves running and even then the messages come in so that slave 1 is almost always used. (By the way, that is better explained in the MAINTAINING STATE section than the GLOBAL VARIABLE LIFETIME section, IMHO). So then, like the MAINTAINING STATE section says, I can put a file in the working directory and test for its presense to decide what to do, right? Then the only change in my solution to the _originally_ asked MX problem is to write out a file to $CWD in the filter_relay() routine indicating that the MX was used (instead of using the RELAY flag from before) and check for that file in filter_begin(). The original logic remains the same. Troy Carpenter [EMAIL PROTECTED] _______________________________________________ Visit http://www.mimedefang.org and http://www.canit.ca MIMEDefang mailing list [EMAIL PROTECTED] http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
