Today, I got worried about two consecutive changes of the ‘ECC Recovered Errors’ parameter reported by SMART on my desktop machine:
Device: /dev/disk/by-id/scsi-SATA_ST3200827AS_4ND2DF00, SMART Usage Attribute: 195 Hardware_ECC_Recovered changed from 67 to 66
Device: /dev/disk/by-id/scsi-SATA_ST3200827AS_4ND2DF00, SMART Usage Attribute: 195 Hardware_ECC_Recovered changed from 66 to 65
That prompted me to see if there have been any similar events recently. Two lines of shell script were more than enough to reassure me that nothing bad was going on:
grep ECC /var/log/messages | egrep -o 'from [[:digit:]]+' | cut -f2 -d' ' > data
And then with gnuplot:
gnuplot> plot 'data' with lines
I got this:
SMART values graph
