New 16 TB HDD... Problems...

[Skak2000]

I have a new 16TB HDD and i have a lot of problems with it.

How does this SMART rapport look? Good/bad? 

Is Read Error Rate with that big a number normal?
 

 

 

Thanks for your help and time!

[jaslion]

What are the problems?

[AbydosOne]

6 minutes ago, Skak2000 said:

How does this SMART rapport look? Good/bad? 

Is Read Error Rate with that big a number normal?

Use CrystalDiskInfo. Those numbers look like incorrectly-parsed garbage.

[Skak2000]

2 minutes ago, jaslion said:

What are the problems?

When i get to about 3,45 TB use the drive get very unstable.
My brand new Synology DS723 mark the HDD as bad.

When i do a read/Write test the transfere speed drop over time.

[Skak2000]

2 minutes ago, AbydosOne said:

Use CrystalDiskInfo. Those numbers look like incorrectly-parsed garbage.

[Levent]

6 minutes ago, Skak2000 said:

When i do a read/Write test the transfere speed drop over time.

Well HDDs do that. Nothing wrong here.

[kirashi]

On 1/5/2024 at 3:20 PM, Skak2000 said:

When i get to about 3,45 TB use the drive get very unstable.
My brand new Synology DS723 mark the HDD as bad.

When i do a read/Write test the transfere speed drop over time.

While I cannot speak to why your Synology NAS is marking the drive as "Bad" (you'll want to check the logs, or contact Synology for assistance), it's absolutely normal for HDD read/write speeds to slow down as you read/write data to the drive.

 

Why? Because seek times are faster the closer to the centre of the drive platters reverse that - I always get this backwards. Most HDDs, if not all, read from the centre of the platters to the outside edges. Due to the laws of physics, it takes longer for the platters to complete a full rotation the further from the centre you are. (It's the same centrifugal force kids experienced when riding a merry-go-round at the playground - the closer you are to the middle, the less you feel like you're going to fall off.) @tikker did a much better job explaining this below.

[Skak2000]

4 minutes ago, Levent said:

Well HDDs do that. Nothing wrong here.

That make sense, Transfere speed is normal 200-250 MB/sek.
But 3-10 MB/sek. was looking wrong... 

[tikker]

On 1/6/2024 at 12:32 AM, kirashi said:

Due to the laws of physics, it takes longer for the platters to complete a full rotation the further from the centre you are. (It's the same centrifugal force kids experienced when riding a merry-go-round at the playground - the closer you are to the middle, the less you feel like you're going to fall off.)

The platter is a solid body, so a full rotation takes the same amount of time whether you are on the inner edge of the platter or the outer edge. The angular velocity is constant across the platter. That does mean that the linear velocity in e.g. metre per second the platter travels increases outwards. The outer edge of the platter thus zips by faster than the inner edge.

 

Like on your merry-go-round example if you on the outside face your friend on the inside, you don't go "out of sync" with your friend since you both rotate at the same angular velocity and the solid nature of the merry-go-round doesn't allow your angular velocity to be different from your friend's.

[kirashi]

6 hours ago, tikker said:

The platter is a solid body, so a full rotation takes the same amount of time whether you are on the inner edge of the platter or the outer edge. The angular velocity is constant across the platter. That does mean that the linear velocity in e.g. metre per second the platter travels increases outwards. The outer edge of the platter thus zips by faster than the inner edge.

 

Like on your merry-go-round example if you on the outside face your friend on the inside, you don't go "out of sync" with your friend since you both rotate at the same angular velocity and the solid nature of the merry-go-round doesn't allow your angular velocity to be different from your friend's.

Thank you - you did a much better job explaining this than I did. I should have explained it as rotation time being the same, but seek time differing depending on the location of data being read/written on the platters. Superuser does a good job explaining the same too.  

https://superuser.com/questions/643013/are-partitions-to-the-inner-outer-edge-significantly-faster

[Skak2000]

Update. The HDD is not stable. It crash if big amount of data is transfer to it.

I try to transfer 200 GB of video and then 80% is stuck and throw error.
Some of the video files was corrupts/damage and WIndows can no longer registered the drive.
Reboot will make the drive come back.
Have been running write and read test for days now with h2testw.

[tikker]

9 minutes ago, Skak2000 said:

Update. The HDD is not stable. It crash if big amount of data is transfer to it.

I try to transfer 200 GB of video and then 80% is stuck and throw error.
Some of the video files was corrupts/damage and WIndows can no longer registered the drive.
Reboot will make the drive come back.
Have been running write and read test for days now with h2testw.

Is this a new drive? If so, you can always try to open an RMA request with where you bought it to get it warrantied. Every now and then it just happens that a drive arrives dead or problematic.

[Skak2000]

9 minutes ago, tikker said:

Is this a new drive? If so, you can always try to open an RMA request with where you bought it to get it warrantied. Every now and then it just happens that a drive arrives dead or problematic.

10 days new drive, have been using 7 days on debugging it... The other 16 TB drive works perfect.
I have open a RMA request now. 
The strange thing is it works but some times with errors. I fear that something strange with it.

----
I remember back in 2008 when the first case of USB scam started to come out.
1 GB USB stick was sold as 2 GB and 2 GB as 4GB.
The drives look like a 4 GB but all data over 2 GB was corrupt. When the drive hid the "fake" storage the write speed will go down big time.
----
That why I was trying to do h2testw and test all the space of the 16TB drive.
But when the write speed drop from 250 MBytes/s til 2-3 MBytes/s. After 3,4 TB I know something must be wrong. And I need to proper test it.
There more data in the drive = more unstable.

I have been working with IT for 15 years, but have not done much hardware testing for long time... I don't know how to read the CrystalDiskInfo.

I was hoping some one was able to see if there is a problem with the drive.

[tikker]

20 hours ago, Skak2000 said:

10 days new drive, have been using 7 days on debugging it... The other 16 TB drive works perfect.
I have open a RMA request now. 
The strange thing is it works but some times with errors. I fear that something strange with it.

----
I remember back in 2008 when the first case of USB scam started to come out.
1 GB USB stick was sold as 2 GB and 2 GB as 4GB.
The drives look like a 4 GB but all data over 2 GB was corrupt. When the drive hid the "fake" storage the write speed will go down big time.
----
That why I was trying to do h2testw and test all the space of the 16TB drive.
But when the write speed drop from 250 MBytes/s til 2-3 MBytes/s. After 3,4 TB I know something must be wrong. And I need to proper test it.
There more data in the drive = more unstable.

I have been working with IT for 15 years, but have not done much hardware testing for long time... I don't know how to read the CrystalDiskInfo.

I was hoping some one was able to see if there is a problem with the drive.

Yeah just send it back in that case. It's not worth the hassle generally in my opinion. CrystalDiskInfo is a good start for sure. Checking never hurts.