On Tue, Nov 04, 2003 at 06:33:49PM +0100, Simon Vallet wrote:
> I have a 4.24 exim server running on a Debian box. It listens on port
> 25 and runs fine. However, when I try to add a second daemon, listening
> on port 465 (SSMTP), it seems the two don't cooperate very well...
> Here is what I get in the SSL-listening daemon logs:
> unable to set gid=8 or uid=1000 (euid=8): local delivery to xxx
> (xxx@???)
> failed to read delivery status for xxx@??? from delivery
> subprocess
> appendfile transport process returned non-zero status 0x0100: exit code
> 1
> The command-line options used are:
> -bd
> -oX 465
> -oP /var/run/exim/exim-ssl.pid
> -tls-on-connect
> -C /etc/exim/exim-ssl.conf
> the binary is setuid root , and compilation options make it run as the
> 'maill' user.
> I tried to add the alternative config file to CONFIGURE_FILE, but it
> didn't seem to help...
> If I understood the way exim is delivering, it actually invokes itself
> to deliver messages locally : could it be that somehow it gets
> confused by the first server running ?
If you want to run two completely different processeces under
different user-ids (maill and mail) you'll have to give them separate
spool and log-directories and set exim_path correctly.
cu andreas
