Hard Drive Errors

[stingray060]

Hi, 

 

I have a Seagate Iron Wolf 2TB NAS Drive, which i have had since December 2018. I don't think it has been abused as such. These are the attributes, obviously they aren't squeaky clean, but how risky is it to run this drive in my server?

 

It seems a shame to throw away a drive!!

 

 

 

 

[trurl]

Is this a disk in your array? I wouldn't be comfortable with that in my array. In order to reliably rebuild every bit of a disk, it must be able to reliably read every bit of all other disks. If at some point in the future you have another disk that needs to be rebuilt, that disk will be involved in recovering that data.

[stingray060]

1 minute ago, trurl said:

Is this a disk in your array? I wouldn't be comfortable with that in my array. In order to reliably rebuild every bit of a disk, it must be able to reliably read every bit of all other disks. If at some point in the future you have another disk that needs to be rebuilt, that disk will be involved in recovering that data.

Ok, So basically, any errors on any drive, is basically a no no. I suppose the risk to cost ratio, it is cheaper, and easier just to replace the drive.

 

I had a look on Seagate's website, looked up the S/N and it seems it is covered under warranty until November this year, so might see if they'll cover this one!

[Hoopster]

25 minutes ago, stingray060 said:

Ok, So basically, any errors on any drive, is basically a no no

A high number of reallocated sectors, especially on a hard drive, is not good. 

 

Here is a very good description of what that means:

Reallocated sectors are bad sectors of the hard drive that can no longer be trusted to safely store data. 
When a hard drive encounters a read/write/verify data error, it marks that sector “reallocated” and transfers 
the data to another spare sector to prevent the risk of data corruption.

 

If there were fewer than 5 and the number was not increasing, you might be able to trust the drive for a while.  Over 1,000 is definitely not good and an indication that the drive is likely failing.

 

I once had a drive with 2 reallocated sectors for over a year and a half.  It never changed and had no other errors.  I left it in the server but watched it closely.  When I saw reallocated sectors jump to 10, I immediately replaced it for the reason Trurl mentioned.  If I needed to rebuild another drive, that "bad" one would be needed in the process.

Edited February 2, 2021 by Hoopster

[Hoopster]

27 minutes ago, stingray060 said:

Ok, So basically, any errors on any drive, is basically a no no

This also depends on the error.  I have a drive in my server currently with 1109 UDMA CRC errors.  These are caused usually by cabling/port issues.

 

Ever since I fixed that the errors stopped and the count has not increased.  The value is stored with the drive and cannot be reset; however as long as it stays at 1109, I have no concerns.

Edited February 2, 2021 by Hoopster

[stingray060]

2 minutes ago, Hoopster said:

This also depends on the error.  I have a drive in my server currently with 1109 UDMA CRC errors.  These are caused usually by cabling/port issues issues.

 

Ever since I fixed that the errors stopped and the count has not increased.  The value is stored with the drive and cannot be reset; however as long as it stays at 1109, I have no concerns.

Ah yes, That explains it.

 

I knew there was a figure on there that, its not crucial if it has errors or not, just that it doesn't increase.

 

Thanks for the knowledge.

P.S. I updated my Signature!!