Pkg exim4 users - Mar 2010 - does failed mean failed?

Thanks for the tip on the regex.  For the record, I was able to define a
new value for the macro in main/00_local, so no need to touch the
shipped config files.

# invoke-rc.d exim4 reload
Reloading exim4 configuration files: exim4.
ALERT: exim paniclog /var/log/exim4/paniclog has non-zero size, mail system
possibly broken ... failed!

config.autogenerated is regenerated, with the new values, despite the
message.  exim seems to have restarted; its log shows
2010-03-21 13:09:29 pid 1299: SIGHUP received: re-exec daemon
2010-03-21 13:09:29 exim 4.71 daemon started: pid=1299, -q30m, listening for
SMTP on port 25 (IPv6 and IPv4)
2010-03-21 13:09:29 Start queue run: pid=14750
2010-03-21 13:09:29 End queue run: pid=14750


What exactly failed?  Is this more of a warning that something has gone
wrong?

The message does not seem to be in /etc/init.d/exim4
or /usr/sbin/update-exim4.conf.

Ross Boylan

On 3/21/2010 1:18 PM, Ross Boylan wrote:
 > # invoke-rc.d exim4 reload
 > Reloading exim4 configuration files: exim4.
> ALERT: exim paniclog /var/log/exim4/paniclog has non-zero size, mail system
possibly broken ... failed!
>
> The message does not seem to be in /etc/init.d/exim4
> or /usr/sbin/update-exim4.conf.
The "failed" is generated by the invoke-rc.d when the init.d script 
returns non-zero.  (This is why I just call the /etc/init.d script 
directly, btw.)

exim4 is almost certainly running ok, but check the messages in 
/var/log/exim4/paniclog to see if there''s a real problem (when this 
happens to me it''s always just spamd timing out on a big/complicated 
message.)  Then null out paniclog (cat /dev/null 
 >/var/log/exim4/paniclog) to keep the message from coming out next time.

Rich