Seagate - Raw Read Error Rate

[DiWiT]

Hi,

I have a 1TB Seagate ST31000528AS and the SMART "Raw Read Error Rate" parameter goes up and down very often, and DiskCheckup reports that it will fail soon. But then, the value changes abruptly and it reports no failures predicted.

Can someone please explain how to understand those values and, most importantly, is my disk actually in a predicted failure state?








I have another 2TB Seagate disk and its "Raw Read Error Rate" value is much more stable over time:



Thanks!

[fzabkar]

Seagate's Seek Error Rate, Raw Read Error Rate, and Hardware ECC Recovered SMART attributes:
http://www.hddoracle.com/viewtopic.php?f=6&t=89&p=13813
http://www.users.on.net/~fzabkar/HDD/Seagate_SER_RRER_HEC.html

Seagate SMART Attribute Specification:
http://t1.daumcdn.net/brunch/service/user/axm/file/zRYOdwPu3OMoKYmBOby1fEEQEbU.pdf

Normal SATA SMART Attribute Behavior (Seagate):
http://t1.daumcdn.net/brunch/service/user/axm/file/Vw3RJSZllYbDc86ssL6bofiL4r0.pdf

[DiWiT]

Thanks,

so I understand that it is not something to worry too much about, but still curious as to why this same attribute on other Seagate disk does not go up and down so often...

Also, this disk came from an iMac and just found out that Apple did a recall and lots of iMacs had to have their 1TB Seagate disk replaced and this was one of the Seagate model numbers affected... uhm, that does not make me feel better, although they said Affected revisions are AP24 and AP25 and mine is AP63.

[Alexey]

I recall, as far as specification goes, as long as normalized value is above its corresponding threshold, the drive is OK, and that's that. The specification does not address changes in values and rate of change. So the question actually is "why do not all attributes jump all over the place all the time". The answer is that behavior of attributes above their corresponding thresholds is implementation-dependent, and Seagate decided to put in little bit less averaging and make their reporting little more sensitive and thus more twitchy.

[DiWiT]

I see there is no clear answer here, so as I use that disk for downloads and the whole contents of the disk are sync'ed overnight to another disk (the 2TB one in the system), I will keep it until it dies (if it ever does, because one never knows).

But one thing is clear after 2 other Seagate disks failed on me in the last few months: next time I buy an HDD, it will be HGST for sure (I didn't actually buy these disks, they were taken from other systems that didn't previously belong to me).

Thanks!