Greetings dman On 01-11-30, dman wrote:
> On Fri, Nov 30, 2001 at 12:38:57PM -0400, cmasters wrote: ... snip a few bits ... > > Hmm, interesting. Do you have gads of stuff in > /var/spool/mail/<localuser> ? That would also explain why stuff that > wasn't directed at your username were lost -- there was no local name > "[EMAIL PROTECTED]" to deliver to (in /var/spool/mail). > At the time, the configuration wasn't recognizing the existence of my standard spool mbox. No idea why not, as I had been receiving mail there during my first days of using 'getmail' before creating an mbox in my home directory. All the mail was 'bounced' back to me in the form of inclusions in the syslog(s), which I received in my $HOME mailbox. > | Apparently when using exim as an MTA, the 'mda procmail' variable in > | 'fetchmail'needs to have '-d <recepient>' flagged, whether is is > | respected or not. Changing the transport instructions in exim seems > | to have the effect of saying "I said to use procmail, now don't ask > | me again!". > > Well, sorta, fetchmail uses the local MTA (via SMTP by default). > Then it is up to that MTA (exim) to deal with it (use procmail for > final delivery). > Now that makes sense. However, fetchmailconf inserts the 'mda ... procmail' line without any flags, which then seems to be ignored (in this default form) as no local user is specified. The man pages (from www.tuxedo.org) don't use ~any~ examples using the 'mda' variant. All in all, that particular man page is ~very~ heavy on technical details (ports, protocols, etc.) and assumes the use of sendmail. > > As I said last night, I would check what fetchmail does for me : > > $ fetchmail -v -v I've changed my cronjob to the above -- from simply "fetchmail" and will definitely post those details for comparison. > > 'dman' is my local username, and I have mda="/usr/sbin/exim %T" in > my .fetchmailrc, which is why it says > "about to deliver with: /usr/sbin/exim dman". I have no idea what the > "About to rewrite" stuff is for. > > > Just some more data for you ... > Very helpful data it has been. > -D C. Masters