Hard Disk Monitoring

I've got (another) HDD with issues but this one has been stable with no decrease in heath since I installed HDS (reporting 20 bad sectors and some weak sectors).

Recently the backup for this drive failed do to the inability to read a sector with the following eventviewer msg: "The backup operation has completed with warning(s) - Volume 'C:' has developed new bad clusters. This may be an indication of problems with your hardware. 1024 bytes have not been backed up because they could not be read. Please run the CHKDSK /R command on 'C:' and then rerun the backup operation." I ran CHKDSK /R on the headless server and post that the backup worked fine.

Getting more proactive I then used HDS to run the HW Quick Test (previously ran fine) and then the Extended Test but they now both fail quickly (Test Failed by Read Element). As per the FAQ I then moved onto the SW test on this disk and selected the Surface Test --> Read + WRITE + read test.

I've only got one bad block so far on the "Disk surface map" but at that time I got a bunch of the following: --> Error: 1117 The request could not be performed because of an I/O device error., Sector: XXXXX

In particular I'm presuming these are related to reading the one bad block marked so far and the HDD will just mark this one not to use and that the disk will then be stable again....but I could not find much on this error and any advice on the HDS forum, any ideas, links or comments on this one would be great.

Thanks
Nathan

The problem is that chkdsk does not do what you expect: it does not force the hard disk to re-allocate the bad sector, to prevent its later use.

Instead, chkdsk marks the cluster of the logical drive (the partition) as bad, forcing the operating system to ignore that area.
But this does not fix the problem, for example if you will ever need to re-partition / re-format the hard disk, the problematic area will be used again - risking data loss.

Exactly as you can see
- the hardware self tests show error
- the Disk -> Surface test -> Read test shows the proper location on the disk surface, indicating that the problematic area is still accessible (not re-allocated)

In this situation (after complete backup) the drive may be placed a secondary drive and Disk -> Surface test -> Reinitialise disk surface test is the best method to repair such problems (the Surface Test --> Read + WRITE + read test is designed for other purposes, not really for this).
This really forces the drive to re-test the weak sectors and
a) re-allocate them if required (to really prevent their future use)
b) stabilize their status by checking if the sectors can be usable now. This may happen if previously a special event (power loss / reset / etc..) caused the weak sector and the problem may not be really related the hard disk.

After this, performing the Disk -> Surface test -> Read test should show no errors, confirming that the hard disk status is now stable. The health% may also increase and creating a new partition on the drive should also show 0 kbytes bad.


In general, the error code 1117 just indicates that it is not possible to process the displayed sector. This and error 23 (CRC error) are the most common results when the software tries to process a sector which does not work as expected (cannot be read/written, depending on the test type started).

Thanks -all makes sence. Followup Q: I get that Reinitialise disk surface test is the best method to repair such problems but the benefit of the Surface Test --> Read + WRITE + read test is that it (designed) to keep the data intact but will it also force the re-allocation of bad sectors?

The "Reinitialise disk surface test" performs overwrite of the drive surface with special combinations, which is the effective method to force the re-allocation of bad sectors.

In contrast, the "Read + WRITE + read test" tries to read the sectors first and if they could be read, it performs the rest of testing (overwrite, write original data back, read back and compare with original contents). Because of its nature, it may not be able to re-allocate the bad sectors, but it may be able to reveal data communication errors and/or if the disk (especially for flash devices) can't keep stored information correctly.

Please check the description of each tests before starting them, the term
"Forces the analysis of any weak sectors and verifies any hidden problems and fixes them by reallocation of bad sectors"
is displayed for Reinitialse disk surface and the simpler Write, Write + Read tests.

The "Read + WRITE + read test" may help, but the "Reinitialise disk surface test" is more effective in this area.

jmone wrote:Thanks -all makes sence. Followup Q: I get that Reinitialise disk surface test is the best method to repair such problems but the benefit of the Surface Test --> Read + WRITE + read test is that it (designed) to keep the data intact but will it also force the re-allocation of bad sectors?

You got this disk error mostly because the source disk has bad sectors . Check the log (see it under the Log tab) to see which partition it was trying to read at the time the error occurred, then try running chkdsk /f against that partition. If it failed while trying to image a hidden partition, you'd need to assign that partition a drive letter temporarily. If it still fails after "chkdsk /f", try the more in-depth chkdsk /r . It is important to note that running CHKDSK with the /r option can take a very long time because it scans every cluster within the partition. If it still fails even after the test, then your disk may be going bad.

http://net-informations.com/q/mis/disk.html

Excuse me, but it seems you may not read what I wrote above....
Please see the message which starts with "The problem is that chkdsk does not do what you expect"

Chkdsk /f (or even /r) does not help in this case.
It would mark the sectors at the level of the FILE SYSTEM (the current partition) as bad, which is not the solution because

1) this just prevents reading/writing the appropriate sectors while the current partition is there. As soon as you re-partition/re-format the drive, the problematic sectors will be re-used again and again, always risking data corruption / data loss
2) when you attempt to clone a such drive, the cloning software may fail as "bad sectors" present
3) if the cloning can be somehow possible, then the partition(s) cloned on the new drive will show bad sector(s) again (even if the drive is perfect)
4) the usable capacity decreases (as chkdsk will report the amount of space used by bad sectors).

So this is not the solution: the solution is to force the hard disk (or SSD) itself to perform reallocation: repair the proper sectors by replacing them from spare area (which is called reallocation).
This would help - as then the original bad sector will be never re-used any more (regardless of re-partitioning, re-formatting or so) and then any logical partition(s) created will not show bad sectors at all - which confirms that the disk drive can be used and even cloned (assuming that the status is stable, so after such reallocation, there are no further problems found).

For this, Disk menu -> Surface test -> Disk Repair is the best possible solution. If I want to be simple, then it performs a more effective function what chkdsk /r would do - just NOT on the actual file system (partition) level but the level of the physical disk drive.

Or if this is "not enough" (if there are really high number of problems), then the Disk menu -> Surface test -> Reinitialise disk surface test is the best choice (simpler functions may be incorrectly called as low-level-format in other tools).

If you are interested, please check:
1) about bad sectors:
https://www.hdsentinel.com/faq.php#health

2) about how to repair weak sectors (which are reused, so usually result that "bad sectors" reported by chkdsk) and how to REALLY fix them
https://www.hdsentinel.com/hard_disk_ca ... ectors.php