Is my HDD on the way out ?

peterworks

Ye Olde Bowler
I appreciate this topic has been covered a number of times but I have been unable to find a reference to this particular problem.
The disk is a Seagate Pipeline HD 5900.2
I ran a Fixdisk -l today and it took nearly 6 hours. The last three took approx 3 hours with roughly the same amount of data. The disk however passed with no apparent problems (log attached).
Looking at the disk diagnostics I noticed that the raw value in '195 Hardware_ECC_Recovered' was growing quite quickly:
195 Hardware_ECC_Recovered -O-RC- 102206103 041 037 000
Then 4 minutes later:
195 Hardware_ECC_Recovered -O-RC- 102226923 041 037 000
Looking at my other box it doesn't even show line 195.

I have been having the odd break up and a couple of freezes when viewing recordings recently and the disk has 'rumbled' for months now. Could this be the end for the disk ?
 

Attachments

  • fix-disk.0.log.txt
    5 KB · Views: 6
OP
peterworks

peterworks

Ye Olde Bowler
Thanks Martin - output attached as screen prints.
 

Attachments

  • Disk Information.jpg
    Disk Information.jpg
    82.9 KB · Views: 12
  • Attributes.jpg
    Attributes.jpg
    170.6 KB · Views: 12
  • Self-test logs.jpg
    Self-test logs.jpg
    178.2 KB · Views: 9

MartinLiddle

Super Moderator
Staff member
output attached as screen prints.
I don't see anything to be concerned about. The Seagate ST31000424CS does have the slightly disconcerting trick of fixing bad sectors but not reporting them as reallocated (maybe it retests them and decides they are OK). The Hardware_ECC_Recovered is exactly the same raw value as the Raw_Read_Error_Rate and I would say that it is not a concern; it just indicates that things are working properly.
 

MikeSh

Well-Known Member
Intriguing. I googled this parameter and arrived at a 2009 discussion which suggested it isn't a problem kind of thing usually, and that a high value is better. There was also a link to an Acronis page which contained this:
Hardware ECC Recovered S.M.A.R.T. parameter indicates time between ECC-corrected errors.
If it's, say, in milliseconds that would make some sense. One would expect it to be updated occasionally and therefore it might go down as well as up.
Or maybe it's something else entirely :oops:
 
Top