what test to run on used purchased hdd

[janzaldua]

hello, i just purchased the PRO version of Hard Disk Sentinel, loving it so far! thanks for the great product.

i had a question regarding buying used drives. I just bought a used Ironwolf Pro 14TB HDD from eBay. What is the best tests I should be running to make sure the drive is ok? The first test I ran was a self-test, and now I'm running the WRITE + read test (which says its going to take 60 hours).

thanks for the help!

[hdsentinel]

Thanks for your message and selecting Hard Disk Sentinel Pro.

The tests you use are absolutely correct: by using the combination of the self-test (hardware self test) and the surface test (especially the Write+Read test) you verify the disk drive functionality in all aspects:
- the hardware self test verifies the "internal" general functionality
- the Write+Read test verifies both the write and read functionality, performance, data transfer - plus verifies that the written data is correct (not damaged/altered/corrupted in any ways)

Generally the Support -> Frequently Asked Questions page has this question:

https://www.hdsentinel.com/faq.php#tests

Hard disk health is low or recently changed or I just installed a new (used) hard disk. How can I perform a deep analysis?


It suggests the following tests:

1) Disk -> Short self test
2) Disk -> Extended self test
3) Disk -> Surface test -> Read test
4) Disk -> Surface test -> Reinitialize disk surface


but on high capacity drives, these (especially the last one) may take really long time, so then it is completely enough to use only the Short or Extended self test and also the Disk menu -> Write+Read test you started.

If these show no errors and the disk health remains high (ideally 100%) then you can be sure that it can be used without problems.

[janzaldua]

but on high capacity drives, these (especially the last one) may take really long time, so then it is completely enough to use only the Short or Extended self test and also the Disk menu -> Write+Read test you started.

If these show no errors and the disk health remains high (ideally 100%) then you can be sure that it can be used without problems.

A few follow up questions.

1. Does this process still apply for SSD and NVMe drives? Do I still run the WRITE+read surface tests on those?

2. Is it fine to test the drives from an external enclosure, or is it best to install them internally first?

3. Since the surface tests take a long time, I might leave a spare laptop on. What's the best way to de-authorize my current install on my desktop, so I can install HDS Pro on my laptop? Or am I allowed to have multiple installations on different systems.

Thanks again!

[hdsentinel]

> 1. Does this process still apply for SSD and NVMe drives? Do I still run the WRITE+read surface tests on those?

Yes, you can perform the tests of course on SSDs (both SATA, SAS, NVMe, USB).
You can also use these tests on pendrives, memory cards or other similar storage devices too to verify/confirm .

Generally a single pass write test is not too much for any SSD, does not cause too much wear - but if you want to minimise writes, then you may use the hardware self tests and the READ test only (without the WRITE test).
Just please note that a single READ test (on a new SSD) may be misleading: as the SSD will likely provide the "data" from the cache instead of the real sectors. This can cause false assumptions (that the SSD seems lightning fast when empty but not later when it actually stores data).


> 2. Is it fine to test the drives from an external enclosure, or is it best to install them internally first?

You can perform the test in any ways: you can use the external enclosure or install as internal drive - as you prefer.
Usually I recommend to test the drive in the place where it will be used: as that way you can verify/confirm that the actual, real operating conditions are good enough (no problems with cables, connections and the temperature level remains in the acceptable range). So by the tests, you do not only test the actual hard disk drive (or SSD) but also the whole storage subsystem to ensure stability.


> 3. Since the surface tests take a long time, I might leave a spare laptop on. What's the best way to de-authorize my current install
> on my desktop, so I can install HDS Pro on my laptop? Or am I allowed to have multiple installations on different systems.

Not sure about your license. Generally all licenses allow lifetime use and free updates to all future versions - but if you have a Hard Disk Sentinel Pro single PC license (not a Pro Family one) then you can use on one computer at any time.
You can simply uninstall and install on the other system if you prefer of course.

Alternatively, you can upgrade your license to Pro Family version so then Hard Disk Sentinel Pro can be installed and used on multiple computers too.

If you prefer, please send an e-mail with your license code to info (at) hdsentinel (dot) com as then I can check the license type and send details about possible upgrade - so you can decide if this may be better solution.

[BOWi_QC]

I feel like I have so many questions about SSDs. I don't even know where to start...

Well... Staying on this topic, I guess I would like you to clarify your stance on testing,

Generally a single pass write test is not too much for any SSD, does not cause too much wear

Does it mean you don't feel comfortable doing a "Reinitialize disk surface" test, or two custom write-read pattern tests?

Just please note that a single READ test (on a new SSD) may be misleading: as the SSD will likely provide the "data" from the cache instead of the real sectors. This can cause false assumptions (that the SSD seems lightning fast when empty but not later when it actually stores data).

What do you recommend we do in order to avoid such misleading test results?

Thanks for such awesome software! (I'll probably create another thread later about SSD and SMR, but concerning older drives.)

[hdsentinel]

> Does it mean you don't feel comfortable doing a "Reinitialize disk surface" test, or two custom write-read pattern tests?

I completely agree you: personally for intensive testing, I also use "Reinitialize disk surface" test, or maybe two custom write-read pattern tests (one with a random pattern and then one with the default 00 fill).

Some users prefer to minimise writes - so they prefer to use as few write cycles as possible.
But generally if an SSD reports problems during the first testing - then it should be immediately replaced during warranty. If the initial testing (which should be generally an intensive testing compared to the normal use) shows no issues - then it is safe to use the SSD later.


> What do you recommend we do in order to avoid such misleading test results?

This is why I personally use the above mentioned methods:
1) "Reinitialize disk surface" test
or
2) Write+Read test, with random pattern (to verify the actual complete performance when real, non-zero data written on the SSD) and then to clear up, the Write+Read test with the default "zero-fill" setting.

The Write+Read test can give better picture about the real performance (as it performs a single-pass overwrite only).
The Reinitialize disk surface performs multiple overwrite passes, so the MByte/sec is lower in this test but it can also show the drop of the performance (with darker green blocks) when the cache filled.

[BOWi_QC]

Thanks!