I've been reading my system logs over the last few days and they're full of entries like this:

May 20 06:26:05 samhain exim[3441]: 2007-05-20 06:26:05 1Hpibr-0000tV-00 Cannot 
open main log file "/var/log/exim/mainlog": No such file or directory: euid=8 eg
id=8
May 20 06:26:05 samhain exim[3441]: exim: could not open panic log - aborting: o
riginal error above
May 20 06:26:05 samhain /USR/SBIN/CRON[3081]: (root) MAIL (mailed 350 bytes of o
utput but got status 0x0001 )

I know exim is a MTA, but frankly I don't even want to try to set up a mail server on my linode (yet); I don't want to even risk making my Linode an open relay because I don't know anything about mail server. But, I would like to be able to setup logwatch and other monitoring daemons to e-mail me reports of the more critical log entries so I know what's going on. I'm fine with receiving those logs via console internally.

I guess my questions are:

1. How do I set up exim to stop the above from happening?
2. What do I need to do, as far as MTAs, etc go to set up the monitoring I've described?

Pointers would be greatly appreciated.

As far as I can tell, Exim is just having trouble getting to the log file specified in its config. First of all I'd make sure that it exists. Secondly, make sure that the user Exim is running as has access to it.

As far as your worries about being an open relay go, your best bet would be to set Exim up so that it only listens on 127.0.0.1. I'm a postfix guy, but as far as I can tell, adding " dc_local_interfaces='127.0.0.1' " to exim's conf file and running "/etc/init.d/exim reload" should do the trick.

Alternatively, the documentation at http://www.exim.org/howto/relay.html has all the info you need on configuring relay access.

Monitoring could be a script of some sort. Install the mailx tools with "sudo aptitude install mailx". That will give you the "mail" command. Then, something like the following should be able to send them.

#!/bin/sh

grep $critical_string /path/to/log | mail -t user@domain.com -s subject