CentOS - Jun 2007 - SMART errors Hardware_ECC_Recovered and Raw_Read_Error_Rate

Hi All,

Im seeing these errors (below) between 2 and 40 times a day, 
for the last year (or more if i had the older logs) in 
/var/log/messages.
The values consistently toggle between 4 or so values, 
drifting up and down but never outside the mean +-2
/dev/hda sits around 21 to 38 degrees (according to smartctl)

Does anyone know *how* serious they are? I cant find much on 
the web other than the theory that any error is bad, so 
replace the disk.

Thanks for your time,

MrKiwi.

Jun 13 04:45:03 myserver smartd[3047]: Device: /dev/hda, 
SMART Prefailure Attribute: 1 Raw_Read_Error_Rate changed 
from 60 to 61
Jun 13 04:45:03 myserver smartd[3047]: Device: /dev/hda, 
SMART Usage Attribute: 195 Hardware_ECC_Recovered changed 
from 60 to 61