When to replace a drive
When to replace a drive
I have a drive that is rated at 92% on the health scale. Generally speaking, at what health rating should a drive be removed from service and why?
-
hdsentinel
- Site Admin
- Posts: 3021
- Joined: 2008.07.27. 17:00
- Location: Hungary
- Contact:
Re: When to replace a drive
In general, under 50% health (this is the "warning" level, when the health turns from green to yellow) it is recommended to consider replacement to allow enough time for backup and replacement.
And under 25% health (this is the "alert" level, when the health turns from yellow to red) it is recommended to perform the replacement as soon as possible.
Usually high(er) health values (90% or above) are fine. It is recommended to have constant monitoring even at this level to be notified about any new problems, any changes in health. If the health is high (90% or above) and the drive status is stable (see below), the drive usually can be used without problems.
Hard Disk Sentinel both checks the errors reported by the drive independently and also checks the relationship between them. It means that if more (different) problems found (usually a sign of near failure) the health degrades very quickly. This is why it is important to take action on these low(er) health levels.
Note that the manufacturers may allow more errors before warranty replacement: as the software indicates, on high(er) health values, warranty replacement may not be possible (you can ask manufacturer/vendor for warranty replacement only on 0% health).
Also it is good idea to verify if the drive has no further (currently hidden) problems, for example on less frequently used areas.
More information: http://www.hdsentinel.com/faq.php#tests
These tests may reveal further problems and (if possible) repair them.
And under 25% health (this is the "alert" level, when the health turns from yellow to red) it is recommended to perform the replacement as soon as possible.
Usually high(er) health values (90% or above) are fine. It is recommended to have constant monitoring even at this level to be notified about any new problems, any changes in health. If the health is high (90% or above) and the drive status is stable (see below), the drive usually can be used without problems.
Hard Disk Sentinel both checks the errors reported by the drive independently and also checks the relationship between them. It means that if more (different) problems found (usually a sign of near failure) the health degrades very quickly. This is why it is important to take action on these low(er) health levels.
Note that the manufacturers may allow more errors before warranty replacement: as the software indicates, on high(er) health values, warranty replacement may not be possible (you can ask manufacturer/vendor for warranty replacement only on 0% health).
Also it is good idea to verify if the drive has no further (currently hidden) problems, for example on less frequently used areas.
More information: http://www.hdsentinel.com/faq.php#tests
These tests may reveal further problems and (if possible) repair them.
Re: When to replace a drive
Hi,
Just bought HDS and one of my 1.5TB disks is showing health of 72% at the moment. You mention waiting till 50% before starting to replace, I wanted to check how available free space should effect this decision? The drive currently has less then 20GB free (it's offline doing the surface scan at the moment
), is it generally better not to over fill less healthy disks do that more space is free for bad sector relocation?
Just bought HDS and one of my 1.5TB disks is showing health of 72% at the moment. You mention waiting till 50% before starting to replace, I wanted to check how available free space should effect this decision? The drive currently has less then 20GB free (it's offline doing the surface scan at the moment
), is it generally better not to over fill less healthy disks do that more space is free for bad sector relocation?-
hdsentinel
- Site Admin
- Posts: 3021
- Joined: 2008.07.27. 17:00
- Location: Hungary
- Contact:
Re: When to replace a drive
Thanks for your message and thanks for using registered version of the software.
In general, the free space does not really matter when we speak about hard disk reliability. It can be filled almost completely - the amount of bad sectors does not decrease the usable capacity of the drive, a special spare area is reserved (which is not accessible until it is required).
Some bad sectors can even cause serious data loss (for example if they are related to Master boot record, boot sector, MFT area or just a critical file), even if their size is very small compared to the total capacity of the hard disk.
The problem is that in some situations the drive status may not be stable which means that further new problems may be found during normal usage. In conrast, it is possible that a drive with 72% is stable: all bad sectors are already re-allocated (replaced by sectors from the spare area) which means that the drive can be used without problems - with constant monitoring and immediate backup on any (even minor) new problem.
The 50% can be used as a guideline, but I know companies (where critical data stored) replacing even at 99% as they can't tolerate any kind of problem which can lead to data loss.
In the current situation, I'd recommend to backup of important files and performing deep tests offered by Hard Disk Sentinel.
These will show if the current status is stable or if there are further problems with the drive and by the tests it is possible to correct the problems by forcing the drive to re-allocate bad sectors, preventing their use in the future. In this situation, it is possible to "clear" the problems and restore its health, but first I'd suggest intensive testing as recommended on: http://www.hdsentinel.com/faq.php#tests
Not sure if you perform surface test with these functions, but I'd recommend Disk -> Surface test methods as these not only verify the disk surface integrity but also reports temperature, transfer rate, retry count issues for particular areas which may show if some specific areas will be likely damaged. For example as shown at http://www.hdsentinel.com/hard_disk_cas ... ectors.php (the end of the hard disk surface has problems)
For more information, please also check:
1) http://www.hdsentinel.com/faq.php#health ( I have bad sectors and my disk health is 90%. Do I need to worry or ask for replacement drive? )
2) http://www.hdsentinel.com/faq_repair_ha ... _drive.php
Just let me know if I can help or assist. Alternatively, you may use Report -> Send test report to developer option (even more than once, eg. after different tests) as then I can check the complete situation of the hard disk and verify how its status may change with time.
In general, the free space does not really matter when we speak about hard disk reliability. It can be filled almost completely - the amount of bad sectors does not decrease the usable capacity of the drive, a special spare area is reserved (which is not accessible until it is required).
Some bad sectors can even cause serious data loss (for example if they are related to Master boot record, boot sector, MFT area or just a critical file), even if their size is very small compared to the total capacity of the hard disk.
The problem is that in some situations the drive status may not be stable which means that further new problems may be found during normal usage. In conrast, it is possible that a drive with 72% is stable: all bad sectors are already re-allocated (replaced by sectors from the spare area) which means that the drive can be used without problems - with constant monitoring and immediate backup on any (even minor) new problem.
The 50% can be used as a guideline, but I know companies (where critical data stored) replacing even at 99% as they can't tolerate any kind of problem which can lead to data loss.
In the current situation, I'd recommend to backup of important files and performing deep tests offered by Hard Disk Sentinel.
These will show if the current status is stable or if there are further problems with the drive and by the tests it is possible to correct the problems by forcing the drive to re-allocate bad sectors, preventing their use in the future. In this situation, it is possible to "clear" the problems and restore its health, but first I'd suggest intensive testing as recommended on: http://www.hdsentinel.com/faq.php#tests
Not sure if you perform surface test with these functions, but I'd recommend Disk -> Surface test methods as these not only verify the disk surface integrity but also reports temperature, transfer rate, retry count issues for particular areas which may show if some specific areas will be likely damaged. For example as shown at http://www.hdsentinel.com/hard_disk_cas ... ectors.php (the end of the hard disk surface has problems)
For more information, please also check:
1) http://www.hdsentinel.com/faq.php#health ( I have bad sectors and my disk health is 90%. Do I need to worry or ask for replacement drive? )
2) http://www.hdsentinel.com/faq_repair_ha ... _drive.php
Just let me know if I can help or assist. Alternatively, you may use Report -> Send test report to developer option (even more than once, eg. after different tests) as then I can check the complete situation of the hard disk and verify how its status may change with time.