2

I have an HDD which was wiped (not by myself).Maybe a program named killdisk was used. I want to re-use the HDD, but it seems to be "really deleted". I don't want any old data from the hdd.

If i start compmgmt.msc and try to initialize a MBR (or GPT) an error Message appears "Data error (CRC check)" If i try it with linux, i can't mount it because there is no "superblock". I can't list it with fdisk -l.

But S.M.A.R.T. Programs can find it and read status information, which are all OK. I used HDDSCAN(win10) and tried to use Tests-->Erase. This function is reporting me "Bad Blocks Found:number". The number is increasing fast. I guess it is full of "bad blocks".

Is this HDD really gone forever? I don't expect that this is possible only by software.

Num     Attribute Name                  Value   Worst   Raw(hex)            Threshold
001     Raw Read Error Rate             118     099     0000000B07-31CD     006
003     Spin Up Time                    100     099     0000000000-0000     000
004     Start/Stop Count                100     100     0000000000-032D     000
005     Reallocation Sector Count       100     100     0000000000-0000     036
007     Seek Error Rate                 079     060     0000010576-C21A     030
009     Power-On Hours Count            094     094     0000000000-15DE     000
010     Spin Retry Count                100     100     0000000000-0000     097
012     Device Power Cycle Count        100     037     0000000000-032D     020
183     SATA Downshift Error Count      100     253     0000000000-0000     000
184     End To End Error Count          100     100     0000000000-0000     097
187     Reported Uncorrectable Error    096     096     0000000000-0004     000
188     Reported Command Timeouts       100     100     0000000000-0000     000
189     High Fly Writes                 100     100     0000000000-0000     000
190     Airflow Temperature             069     052     31 C                045
190     Airflow Temperature Minimum     069     052     24 C                045
190     Airflow Temperature Maximum     069     052     38 C                045
191     G-sense Rate/Servo tracking     100     100     0000000000-0013     000
192     Emergency Retract Count         100     100     0000000000-0015     000
193     Load/unload Cycle Count         090     090     0000000000-55E6     000
194     HDA Temperature                 031     048     31 C                000
194     HDA Temperature Minimum         031     048     16 C                000
194     HDA Temperature Maximum         031     048     52 C                000
195     Error Rate                      046     038     0000000B07-31CD     000
196     Reallocation Event Count        100     100     0000000000-0000     036
197     Current Pending Errors Count    100     100     0000000000-0000     000
198     Uncorrectable Errors Count      100     100     0000000000-0000     000
199     UltraDMA CRC Errors             200     200     0000000000-0000     000
254     G-shock/Free-fall Event Count   100     100     0000000000-0000     000
7
  • 1
    SMART does not detect and display all errors. However, unless you provide the values, we cannot check this.
    – Daniel B
    Commented May 11, 2021 at 18:13
  • Is this a HDD from a professional storage system? If yes, you might need to low-level-format it, as it might have a block size not a power of 2 (520 Bytes is used quite often) Commented May 11, 2021 at 18:14
  • it's from an enterprise PC
    – Stefan
    Commented May 11, 2021 at 18:55
  • 1
    Low-level formatting hasn't existed for many many years. It's simply not possible on IDE (and newer) disk drives.
    – Daniel B
    Commented May 11, 2021 at 19:04
  • @Eugen Rieck, there is no low level format anymore: seagate.com/support/kb/…
    – r2d3
    Commented May 13, 2021 at 22:15

2 Answers 2

0

What your software states on blocks (which is kind of above sector level, as blocks are more like clusters in NTFS p.e.) ...

But S.M.A.R.T. Programs can find it and read status information, which are all OK. I used HDDSCAN(win10) and tried to use Tests-->Erase. This function is reporting me "Bad Blocks Found:number". The number is increasing fast. I guess it is full of "bad blocks".

... is not reflected in the SMART parameters: There are zero pending and reallocated sectors. But your SMART parameters show unusual figures for the seek error rate.

=> It's not bad sectors that cause your bad blocks!

-1

since you had it visible in linux- if you want it to be completely like a new HDD,

dd bs=1024k count=10 if=/dev/zero of=/dev/[whatever it's dev name is]

that'll zero out the first 10 megs of the disk (more than enough) wiping out any previous partition table format, or superblocks. Eugene Rieck is also on point, if the disk is from an exotic enterprise system of some variety, it may be in an alternate block size, which can be resolved via a low-level format.

If the HDD was wiped using a magnetic degaussing tool, it may have also wiped some of the NVM chips on the logic board, losing factory calibration data. That is generally non-recoverable for any but the most capable data recovery services.

3
  • 1
    I tried it. and the output was: 10+0 Datensätze ein(Datasets in) 10+0 Datensätze aus(Datasets out) 10485760 Bytes (10 MB, 10 MiB) copied, 6,16506 s, 1,7 MB/s But it did not change the described behavior. Still can not create any partition.
    – Stefan
    Commented May 11, 2021 at 19:25
  • @gusfoo, there is no low level format, see my link above in the previous comment. Zeroing a drive does not change its sector size.
    – r2d3
    Commented May 13, 2021 at 22:18
  • @r2d3 I did not say a zero would change it's block size, I suggested that to wipe any strange partition table or superblock, that could be confusing a simpler tool like windows. I should have made the two thoughts more distinct from one another. And, for the ATA market, you are right, and low level is no longer a thing. However, in SCSI land, this is most certainly still a thing, as mainframe (zSeries), power systems (AS400, etc..), and such can attest to. Also, on Linux, sg_format with -F and -s, while somewhat dependent on your hw capabilities to work, is not there just for show.
    – gusfoo
    Commented May 15, 2021 at 2:24

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .