Overclock.net banner

HDD SMART Data and Drive Replacement

217 views 7 replies 4 participants last post by  Lord Xeb 
#1 ·
I'm looking at HDD SMART data for some WD Green drives and had some questions on interpreting the results.

I have some drives with over 10,000 Power on Hours and about 2000 Power on Count. That's roughly 500 days powered on... If there are no other errors (e.g. Read Error Rate or Reallocated Sector Count) when should I consider replacing the drives?

Now, one of these drives is showing Read Error Rate RAW value of 44 but no Reallocated Sector Count or any other errors. Is that grounds for a definite replacement? I'm not even sure when these errors occurred.

Just wondered what your guys' thoughts were.

Thanks

P.R.
 
#2 ·
Please post a screenshot of the smart data and i will help you out.

Also, Read Error Rates are normal. Doesn't seem bad at all. Generally speaking, if there are no reallocated, uncorrectable or pending sectors, you are fine. 10k hours is not that bad. I see a lot of drives in my job that have 25-30k (if not more).
I would do a surface scan on it to make sure there are no bad sectors. Occasionally, smart doesn't update till a full surface scan is done. Additionally, do a performance test on the drive while you are at it. If the drive is around 80-100MB/s read then it is doing fine.

A drive will usually last about 3-5 years.
 
#3 ·
Thank you, this was very informative.

Here's an interesting article I found about HDD failure rates which may provoke some discussion.

http://www.extremetech.com/computing/170748-how-long-do-hard-drives-actually-live-for

They state that in their tests the drives are "constantly online" so that makes 3-5 years 26280-43800 hours. Does that sound fair?

My drives seem to only have accumulated less than half of that in the same amount of time so I'm curious, how do other age related factors, such as Power on Count, affect the failure rate? What about workload? (Read/Write).

Is there any info on how much read/write, or on/offs, a drive is rated for?

TLDR of the article: Drives older than a year and younger than 3 have the lowest failure rate.

0-1.5 years 5% of drives fail
1.5-3 years 1.4% of drives fail
3+ years 11.8% of drives fail

Here's a snap of the SMART data but based on what you're saying it sounds okay. (Using Crystal Disk Info)
 
#4 ·
That drive looks really healthy. Also, a lot of that is propiretary so you are not going to gain much access to it. Also, you are talking to someone who works for a data recovery company for a living. >.> That drive looks incredibly healthy for its age. I wouldn't worry about it.
 
#5 ·
I wouldn't depend too much on SMART readings since a drive can suddenly, not to mention irrecoverably, fail without any notice at all. You simply cannot count on any drive, even a new one, to be 100% reliable. The only way to ensure your data is safe is to maintain multiple backups. A general rule of thumb for data protection is to back it up to at least two places, such as an onsite backup, which can be as simple as an external HDD kept at home and connected to the computer only when updating the backup, and an offsite backup. The offsite backup can be an external HDD stored at a trusted friend or neighbor's house, a locked drawer or locker at work, a safe deposit box at a bank, etc. or you could use a paid cloud backup service such as Carbonite.com, Crashplan, or Backblaze (stay away from free cloud storage sites; besides being unsecure, they are notorious for disappearing with little to no warning, taking your data with them). The idea is your data has to exist in at least three different places to be reasonably safe.
 
#6 ·
Hi @PtrRabbit42!

The results don't show anything wrong with the drive - it looks healthy and OK.

Just to be sure you can run some additional tests with our tool called Data Lifeguard Diagnostics. I suggest to run an extended test instead of the quick one because the software is tuned to try fixing automatically any errors it finds, but make a backup of your most important data first. Here's a link to download the program:

http://products.wdc.com/support/kb.ashx?id=8UAb5g

Hope this helps and let us know how it went. Also, feel free to upload some screenshots if you want so I can take a look at the results.
smile.gif
 
#7 ·
Thanks for all the input everyone. I will run an extended test on the drive and watch the SMART data (or enable SMART monitoring).

@Lord Xeb - You said earlier that occasionally SMART doesn't update until a surface scan is done. Will the regular Check Disk do this or does it have to be a deep scan (Scan for bad sectors)?
 
#8 ·
It would have to be a full scan. Your drive would give you warning signs (usually) when it starts dying. I would really not worry about it. Just have a good backup scheme and replace it when it eventally dies. How it is.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top