Improve company productivity with a Business Account.Sign Up

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1413
  • Last Modified:

Hard Disk Failure: read-error timeouts take too long

I have a failing hard disk (Hitachi HDT721032SLA380 - 320G, SATA). I know there are 3000 or more sectors that error on reading.
I'm running ntfsclone --rescue on Linux to make a copy (to a file on an external hard disk), but it reads in 512byte chunks and takes 30 seconds to timeout on each failure. (From what I can see, the errors are in 4096 byte chunks, so it actually fails 8 times (4 minutes) for each failing chunk.)

I've tried changing /sys/block/sda/device/timeout from 30 to 5 but that hasn't speeded anything up (should it?)

Does anyone know a way to reduce the timeout with or without stopping ntfsclone?
0
Beneford
Asked:
Beneford
  • 3
2 Solutions
 
nobusCommented:
i would stop ntfs clone rightaway, and try to repair the drive.
i would connect it to a working windows pc, and run hddregenerator on it, it has saved many drives for me : http://www.dposoft.net/hdd.html      

**but that is only my way of doing things
0
 
BenefordAuthor Commented:
I appreciate the your comment.

Right now, this is a non-cost project, but I may give HDD Regenerator a trial later.
I'm also keen to see which files are potentially corrupted, which ntfsclone is doing (slowly!)
0
 
DavidPresidentCommented:
If this was a SAS drive, then the answer is that this a a programmable setting via a mode page editor software. The SATA spec doesn't let this setting be changed.  This timeout is the HDD going to deep recovery.  It is trying every trick in the firmware to recover those blocks.

Let it run. Nothing you can our SHOULD do about this.  It may take days or a week or longer.  Sorry.

Next time buy ENTERPRISE class SATA.  The recovery will give up in just a few seconds.  But you have a cheap desktop drive that is not designed for 24x7 use, and this is just the "no-free-lunch" downside of buying a consumer class disk drive.

Note also enterprise class drives have more ECC bits, so it is unlikely that you would have more than just a few unrecoverable blocks had you bought one of those drives instead.
0
 
BenefordAuthor Commented:
Thanks, dlethe,

It's nice to know there's no free lunch.

This is a home PC rather than an enterprise system (so no RAID, no expensive disks, no backups(!)). But lost data is sad rather than catastrophic.

Given that I've completed a full copy with dd_rescue, the ntfsclone is additional on the failed partition with the failures, and I'm as interested in the listed "Can't read sector at nnnnn" as in the actual data it recovers.
dd_rescue let me specify the minimum block size which meant only one delay per failing group-of-8-clusters. ntfsclone doesn't, and is doing 512 bytes at a time.
0
 
BenefordAuthor Commented:
Thanks guys for your thoughts.

The ntfsclone failed (returning 100s of errors per second) so I abandoned it.
The dd_rescue I had previously run seems to have worked, so alls well.
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Easily Design & Build Your Next Website

Squarespace’s all-in-one platform gives you everything you need to express yourself creatively online, whether it is with a domain, website, or online store. Get started with your free trial today, and when ready, take 10% off your first purchase with offer code 'EXPERTS'.

  • 3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now