5

2 days ago, during download, uTorrent started to throw CRC error (write), i resumed failing download multiple times with no luck(same error appeared again and again just in few seconds/minutes), then decided to stop it and run full chkdsk.

  • It took about 30 hours to complete for 1TB partition, stuck for hours on few files (see output)
  • During chkdsk i run smartmonctl --health (when i thought chkdsk just stuck on some file), and smartmonctl reported raw_read_error_rate failing now, plus a friendly message hdd will die in 24hrs
  • after chkdsk completed, smartmonctl --health shows PASS and indicates that raw_read_error_rate failed_before (current normalized 93, worth 1, threshold 53)
  • checkdsk output shows ~12000 bad clusters:

C:\Users\Taras>chkdsk E: /F /V /R
The type of the file system is NTFS.
Volume label is Data.

Stage 1: Examining basic file system structure ...
47360 file records processed.
File verification completed.
27 large file records processed.
0 bad file records processed.

Stage 2: Examining file name linkage ...
48220 index entries processed.
Index verification completed.
0 unindexed files scanned.
0 unindexed files recovered.

Stage 3: Examining security descriptors ...
Cleaning up 46 unused index entries from index $SII of file 9.
Cleaning up 46 unused index entries from index $SDH of file 9.
Cleaning up 46 unused security descriptors.
Security descriptor verification completed.
430 data files processed.

Stage 4: Looking for bad clusters in user file data ...
Windows replaced bad clusters in file 2813
of name $PATH1.MKV.
Windows replaced bad clusters in file 2863
of name $PATH2.MKV.
Windows replaced bad clusters in file 2881
of name $PATH3.MKV.
Windows replaced bad clusters in file 2891
of name $PATH4.MKV.
47344 files processed.
File data verification completed.

Stage 5: Looking for bad, free clusters ...
88458934 free clusters processed.
Free space verification is complete.
Adding 12163 bad clusters to the Bad Clusters File.
CHKDSK discovered free space marked as allocated in the volume bitmap.

Windows has made corrections to the file system.
No further action is required.

1347949889 KB total disk space.
993898020 KB in 15398 files.
12752 KB in 432 indexes.
48652 KB in bad sectors.
154729 KB in use by the system.
65536 KB occupied by the log file.
353835736 KB available on disk.

4096 bytes in each allocation unit.
336987472 total allocation units on disk.
88458934 allocation units available on disk.

HDD is external Samsung Story Station 1.5TB 5400 USB2 Green (auto suspend).

The question is: does it necessarily mean close death of disk, or can it be safely used as media storage (no important stuff)?. I have a feeling those errors may occurred because of disk suspends/resumes along with software problems, can it be turned off somehow?

Or should I just thrown it away?

4
  • 2
    Never, ever run the risk and think "I'll do it later"... It always ends in tears. Why don't you have a backup? A good backup strategy and you wouldn't need to ask the question. A mechanical hard disk can fail at any time.
    – Kinnectus
    Jan 3, 2015 at 13:08
  • The bad clusters can be just corrupted file contents but the bad sectors are worrisome (they are defects). That's a lot. You could repeatedly test it to see if new bad sectors keep popping up. If so, toss it. If not, do a low level reformat. Then use it for inconsequential storage with periodic retesting, or as scratch space with periodic retesting. I wouldn't trust it with anything important.
    – fixer1234
    Jan 3, 2015 at 18:38
  • @fixer1234 are that number of bad clusters excluding already remapped? HDDs should have a pool for remapping bad clusters. But i agree that 12 thousands is a lot... Jan 4, 2015 at 9:47
  • Supposedly, /r and /f should move any data it can recover and fix the errors. My recollection is that there is some maximum number of entries in the table it uses to keep track of this.
    – fixer1234
    Jan 4, 2015 at 10:11

2 Answers 2

5

Never trust a failing drive with data you care about. Your drive is failing. By the time chkdsk starts reporting bad sectors, your drive has exhausted all of the reserve sectors it came from the factory with to deal with the small number of "normal" sector failures that occur in otherwise healthy drives. However, once the reserves are depleted, trouble is brewing and the drive is on its way down the tubes.

It's hard to say if the drive will continue its path of self-destruction by reporting more and more bad sectors (corrupting everything in its path along the way), by suddenly refusing to be recognized by your computer, or by some other spectacular failure mode. Regardless, if there's something on that drive you want when this happens, you'll regret having not already moved it to a safer place.

4
  • thanks for an answer, so this number of bad sectors usually excludes already remapped? it's even worse then. it's media drive, no important info, i was surprised that it suddenly reported that bad, as it was used occasionally Jan 4, 2015 at 9:50
  • Yes, when a drive detects and remaps bad sectors on its own, they're not visible to programs like CHKDSK and therefore won't show up on its report. Jan 4, 2015 at 12:50
  • what tool can you recommend to periodically check HDD health (cmd line tools preferably, but not necessarily)? I'd still like to use this HDD until it finally dies, but also want to monitor it Jan 4, 2015 at 13:58
  • I manage too many systems to do much proactive monitoring of individual HDDs health (I rely on RAID's built-in monitoring instead), but when I'm assessing HDD health I always prefer to use the drive manufacturer's own diagnostic tools. For this, I use the Ultimate Boot CD which is free and includes non-OS specific diags for all major HDD manufacturers. When dealing with an unhealthy disk, I turn to GRC's SpinRite to get problems fixed. Jan 8, 2015 at 18:29
-2

This is my first Answer ever on this site: Firstly, just throw that HDD away, but I wanted to discuss other approaches to your common problem:

A CRC error (write) error, or an I/O error, with an external USB HDD, can be that your external HDD write speed is too slow. When an HDD is failing in a RAID 5 array, the write speed oscillating and slow, is often the very first sign of impending failure. Add a USB bottleneck to that, and you will get I/O errors. In a RAID array failing in this way, you can painstakingly one by one, replace the HDDs, until you find the offending HDD, or backup, and wait until HAL throws you out the airlock. Like Windows, sometimes a NAS needs a clean install, which is never pretty.

Using the NTFS format on an external HDD, is a prescription for trouble. NTFS format (and exFAT), tend to fragment files like hell, and also starts file writes 'at the spindle of a floppy drive' and works the file writes outward. This is because of the history of DOS. NEVER use exFAT on an external HDD! exFAT is marvelous, but only use it on mSD memory cards and USB thumb drives -> Solid State.

Ext4 is 'exclusive' to LINUX; I use Ubuntu. Always use Ext4 format on SSD, HDD, including external HDD, if that is a functional option. Ext4 avoids fragmentation, and even on a solid state device, does not write repetitively from 'at the spindle of a floppy drive'.

On a NAS, BRTFS is the highly recommended format, but my advice is use Ext4, and backup, backup, and backup. Most gurus and experts would condemn this advice, but for my 'Mission', I say Keep It Simple, Stupid. Now, BRTFS has a very desirable ability to sometimes repair its files, but I choose to not use that feature. Furthermore, I don't ever use encryption for anything as I simply have physical control of it all.

You are doing torrents, and I advise that if any of those are on an SSD, use an external SSD, to unload the heat.

I suspect that one of the reasons the ISPs have tremendously increased our bandwidths, is to make throughput so rapid, that an external HDD becomes a bottleneck and I/O errors occur.

Large file transfers seem to create memory holes. An Ubuntu app, pSensor, follows temperatures, CPU use, and Free Memory. As Free Memory runs out, there is more CPU use and using of the platform's HDD or SDD. The sure go to way to remedy these is to reboot the platform.

Apple File Protocol, my favorite although I own no Apple stuff, is going or gone bye bye. For me, that leaves SMB, Server Message Block. SMB and exFAT are arguably the very best of alive today Microsoft, although SMB started IBM. exFAT is just another DOS: Disk Operating System. I have one NAS backing up to another NAS, and once started, those two devices need nothing but themselves and a switch (routers usually include switches), to communicate. Cutting out the middleman Ubuntu platform, is marvelous.

There is host of other protocols and issues, and while there is a LINUX native NFS protocol, what the world needs, is analogous to the USB-C charging port on EVERYTHING!: we all need a file transfer protocol which can transfer files all day long, without jazzing up memory swapping, the CPU, write speed bottlenecks, Free Memory, ... ... ...

Very finally, if you had tried an external USB SSD, with a TEMP folder on the USB SSD, a Crucial X8 4TB Portable SSD (the 4TB and 2TB have the bigger cache, while the newest X9 only adds encryption), your I/O error would have disappeared.

BONUS: If UHD/4K playback stalls or glitches on playing from an external HDD, but works and fast forwards etc. from the platform's SSD, the Crucial X8 4TB Portable SSD will resolve that. Just as you cannot file play UHD/4K from a NAS with VLC pulling the file from over an Ethernet cable, well, copy whatever to the Portable SSD, and you will be good to go. You can go 'old school' and sneaker-net from the NAS to the Portable SSD. Voila! You can fastforward, jump around, ... ...

You must log in to answer this question.

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