Qmail to Exim woes
OK, it's my turn. Yes I RTFM's.
First the date. My local network is mtntop.home, my user name is
wtopa (of course), my address for the net should be
I decided that if exim was the prefered Debian MTA I would switch to
it. Well after 3 days I have finally got it te accept mail from the
net but I still can't send any ( I'm on Slackware (qmail) now).
If I follow the eximconfig instructions, option 2. I get the following
qualify_domain = ix.netcom.com
local_domains = ix.netcom.com
and the correct smarthost section.
No mail is received with the above. Fetchmail won't get the mail and
replies with something like "can't accept mail to localhost, no wtopa
Based on that message I changed some of what the script configured
qualify_domain = mtntop.home
local_domains = localhost
I left the smarthost line alone.
Result: I can now receive mail. I still can't send mail so I added
to REWRITE CONFIGURATION
email@example.com firstname.lastname@example.org frs (As per a post to
Attempts to send mail give the following log entries:
email@example.com routing defer (-32): retry time not reached
1999-03-05 11:56:55 10Ixsz-0000JN-00 == firstname.lastname@example.org
routing defer (-32): retry time not reached
1999-03-05 11:55:41 10Ixsz-0000JN-00 <= email@example.com U=wtopa
firstname.lastname@example.org R=smarthost defer (-1):
These log entries were after an exim -q as root.
I did not try using eximon to send these, maybe I should have, but
they should send without it anyway.
Things I tried but that didn't work.
I tried local_hosts = mtntop.home as also suggested but after going
tru all the docs can not find any reference to that so removed it.
So my questions are :
Is the eximconfig script setting exim up for anyone else?
What is required to send mail with this beast?
I have removed the 30m from the init.d/exim scripts as I cron for mail
twice an hour, is this OK?
Any suggestions or flames appreciated.
Real computer scientists despise the idea of actual hardware. Hardware
has limitations, software doesn't. It's a real shame that Turing
machines are so poor at I/O.
Wayne T. Topa <email@example.com>