How to know if my HD has to be replaced?


I am still gathering information, making different tests, since my HD seems to “hold-on” for now. The important data has already been backed up. I just want to make sure before spending my bucks on a new HD that I would not need at all if my current one can survive.

Is there any way to make sure that my HD needs to be replaced ?

Seagate ST380020A (80GB, 5400rpm), 2 years old (5500 hours according to SMART).
No weird sound coming from the HD.
No problem while running Windows (2000) except a blue screen of death that showed up 3 times (in a week or 2) before I decided to start checking around.
Temperature <45 degrees while running full disk scan. Room temp is about 22.

I ran tests with HD TUNE that showed errors:

  • quick scan shows only 1 block damaged (always the same block)
  • full scan (slow scan) showed a series of damaged blocks (7.2%) during the first scan. The scans I did after that showed different damaged blocks (but the scans now show each time the same blocks).

Windows 2000 “chkdsk” found errors, and apparently repaired them (but HD tune still shows error blocks; is it normal?).
BTW, how come chkdsk needs to run at boottime while HD TUNE can perform its scan with Windows running?

The Seagate diagnostic tool for scanning HD fails all the time: quick test fails saying impossible to communicate with the drive (it’s not a “bad block” error). Full test fails for same reason (without doing any scanning, I think). I have the feeling that the tool fucks up (and maybe not the HD).

Hitachi diag tool failed too (right away).

If someone finds it interesting, I can post the log of chkdsk, and the results of HD TUNE; I can run other tools if needed.

Mmm… what should I do?
Thanks a lot! :bow:

If Seatools says it’s bad you should take backup of sensitive data asap and replace it.

As I said, I already backed up the files. The computer that uses this HD is not aboard a space shuttle (if it stops working for a few days that’s no big problem), and my budget is not the one of NASA either.

Do you mean that Seatools is always right? (has no bug?) So they make better software than hardware, uh? (sorry for that one, I am not so happy with Seagate these days… :frowning: )
I find it a bit hard to believe Seatools since it cannot even start its scans, and then gives me a fairly useless error code.

I also think I saw (it was displayed for half a second, then erased by something else) a 0x22 error code.

On Seagate website I can read:
“Defective drive” and “Corrupted format” errors (sometimes called 03/31 errors) can often be repaired with a data-destructive zero fill data pattern or a low level format. Current disc drives contain thousands of spare sectors which are automatically reallocated if the drive senses difficulty reading or writing. Since SeaTools is read-only (data safe), occasionally a problem sector that has not reallocated to a spare sector can be forced to do so by writing to that sector. Spare sector reallocation is a normal intelligent drive operation."

So for example, what about a low level format ?

I also got “AtaCore Error 2 in SeqScan” and “AtaCore Error 11”; their doc is not clear at all about that. They say something like: “try again after restarting the machine. If it does not fix it, replace the drive.” Nothing more precise. :frowning:

I would have to say What’s more important data or $$$. Also download Memtest86 and check your memory. Mem errors can cause hdd errors. But i would have to say if sea tools says its bad buy the drive.

Or… Do a low level format reload windows and try it for awhile. Do a clean install not an image of your old drive. (sometimes partition and file system errors will copy into the image and you’ll have the same results. Run the drive with the newly installed windows and see if you get errors. If not, drive is probably OK and you can restore your data.

Hello again, :bigsmile:

If anyone knows about SMART attributes, I’d love to hear his/her opinion.
I attached 2 snapshots of the numbers given by HD Tune on 2 dates separated by 6 days.
You can see a few parameters changed:

  • Raw read error rate 65->66
  • Reallocated sector count 63->62
    and a bunch of changes in the raw data.

Does these numbers give any clue about how fast the HD is dying (if it is) ? :confused:

NB: I ran several times a full chkdsk (XP) and the result is now always the same (this looks OK I think):
Checking file system on C:
The type of the file system is NTFS.

A disk check has been scheduled.
Windows will now check the disk.
Cleaning up minor inconsistencies on the drive.
Cleaning up 13 unused index entries from index $SII of file 0x9.
Cleaning up 13 unused index entries from index $SDH of file 0x9.
Cleaning up 13 unused security descriptors.
CHKDSK is verifying file data (stage 4 of 5)…
File data verification completed.
CHKDSK is verifying free space (stage 5 of 5)…
Free space verification is complete.

67657715 KB total disk space.
21694776 KB in 122007 files.
41300 KB in 8297 indexes.
12288 KB in bad sectors.
190951 KB in use by the system.
27680 KB occupied by the log file.
45718400 KB available on disk.

  4096 bytes in each allocation unit.

16914428 total allocation units on disk.
11429600 allocation units available on disk.

I gave up trying to run any other diagnose tool; they all fail without giving any clear error message. :pukey:

PS: Please don’t repeat me to backup my data, I got that part all right. :iagree:

your seek/read error count is massive!
you go some 11MB in bad sectors…

get a new HD asap…

Thank you, in deed these numbers worry me, but:

[li] I saw at least one post in another forum where guys were comparing their Smart attributes; one drive had similar error rates, and no one in the thread seemed to be afraid of it (well, maybe they did not notice, I admit)
[/li][li] This numbers are “stable” (at least for more than 10 days)
[/li][li] Why would Smart say it’s OK if this clearly shows the drive will die soon??? (threshold is 25, my HD still has 65-66). What’s the point of Smart, then, if you have to change a drive that shows “OK” status?
Once again, I got the point, though: I cannot trust my HD anymore (could I ever?).

The only important field for u is DATA since it reflects the # of occurrences of a particular error. By this means, u got 397 sectors marked as defect and relocated.
Your UDMA CRC error count is 11 accompanied by a massive amount of read/seek errors.
Of course, any dealer will tell u no problem until the hd is dead. It is discussable if 12MB of bad sectors is significant for a 80 Gig HD. However, it is your data and u have to take precautions (backup in time). HD data (bits) of today’s drives are written in ultra high-density mode meaning bit dropouts can happen much sooner due to an “aging” magnetic layer. It is all up to you…

If Smart Tools is telling you it has errors, I’d send it back to Seagate. Use this link to verify warranty and create an RMA number.

You are responsible for the cost of shipping it to Seagate and they pay to ship your repaired drive or a replacement drive to you. I’ve never had them send me a repaired drive and I have RMA’d probably close to 200 drives with Seagate alone.

Just type in your model and serial number then select the country you live in and it will tell you if the drive is under warranty.

If it is under warranty, make sure to package your drive well. Any damage that occurs in shipping will void the warranty, GUARANTEED.

Sorry I had no time to come back here and thank you before now.

  • My seagate drive was not “returnable”, if I understood correctly, because I did not buy it separately. It came in the standard configuration of a Fujitsu PC, which was more than 2 years old.
  • I did not lose data in the story; the SMART numbers of my drive (reallocated sector count, specifically) just kept on increasing slowly, until I bought another drive (a Maxtor, BTW).
  • I was not able to image-copy my seagate drive to the Maxtor due to some weird Windows problem (Windows XP crashed systematically when detecting the Maxtor on the secondary IDE; a thread exists somewhere about that).

Nothing more to say about this. Thanks all for your help.