Hi,
[...]
> Has anyone else come across this or is it just me?
It's not only you. I wanted to know, why syslogd ate 70% of my CPU time
and saw that it just logs enormous amounts of data to /var/log/postgres.
So the first thing was to put a hyphen in front of the logfile's name in
/etc/syslogd.conf and the second best thing was to set the debug level of
postgres to a more sane value.
After doing this and restarting both syslogd and postgres, the performance
was five to six times better than before. So, maybe we can have a note
about that in the postgres manual?
Kind regards
... Ralph ...