Re: [exim] Trim down pgsql callouts

On 2008-07-09 at 22:03 +0100, Sam Smith wrote:
> So I ran the command exim joseph@??? 2>&1| tee /path/to/logfile

If you're invoking Exim like this and the initial delivery is deferred,
then the invoking Exim has to talk to the DB and later the queue-runner
needs to talk to the DB separately.

Also, lookup caches are flushed after routing, for instance, so if
there's data that you want to use again at delivery (Transport) time, it
makes sense to use "address_data" on the Router to preserve it and
$address_data in the Transport to retrieve it; this data is stored in
the spool file.

If you're normally injecting via SMTP, then can you please run with the
-bd daemonising flag, send the mail via SMTP and post the full debug
output, since (as you say) there's nothing confidential in it?

It's probably worth doing "exim -bd -d+all 2>&1 | tee debug.log" to get
all debugging possible, so that we can pick over it in detail. Please
don't cut out parts of the log -- send the whole thing, since we need to
be able to figure out what's changing to cause Exim to do the lookups.

If you're trying to avoid interfering with an existing production Exim,
then use "-bd -oX 26 -d+all" to listen on port 26.
an existing production Exim, then use "-bd -oX 26 -d+all" to listen on
port 26.

How much of your configuration are you happy posting? If all of it,
then good. :^) You can (probably) strip comments for this audience:

grep '^[^#]' `exim -bV | sed -n 's/Configuration file is //p'` >conf

-Phil