[SOLVED] New WD Red HDD > blk_update_request: critical medium error

[parazit]

Hello guys!

 

I have just purchased a new WD Red Pro 6TB and i am seeing a lot of "blk_update_request: critical medium error" in the pre-clear log.

The disk is connected through a USB Adapter instead of SATA at the moment, because all SATA ports are in use.

Is that a normal behavior?

Edited November 8, 2021 by parazit

[itimpi]

No - sounds like a faulty disk.    You could try running an extended SMART test to confirm.

[parazit]

I have did the extended SMART test and the test was good. After that i ran the pre clear script and everything was fine except of the post read verification. Is that something that i have to worry about?

Also the medium error in the Disk Information Log is still present, but thats kinda weird to me because i have those errors with a disk that failed and which i already replaced.

Any ideas?

 

Logs are attached.

Log.rtf preclear_disk_V9G95D3R_25071.txt

[trurl]

7 minutes ago, parazit said:

post read verification

Have you done memtest?

[itimpi]

1 hour ago, parazit said:

After that i ran the pre clear script and everything was fine except of the post read verification.

Yes - this means that UnRaid cannot read successfully the data that has just been written.

[parazit]

@trurl thank you for your answer. Could you be so kind an tell me how to run a memtest? I made several searches but didn't find a solution how to run a memtest.

 

@itimpi thank you also for your answer. Is it possible to only run a post read instead of performing a complete pre clear? I just want to make sure the disk is fine, so i can investigate where the error is coming from.

[itimpi]

10 hours ago, parazit said:

Could you be so kind an tell me how to run a memtest? I made several searches but didn't find a solution how to run a memtest.

It is one of the options on the Unraid boot menu.  Having said that if you boot in UEFI mode then y/u need to download a UEFI compatible version from the memtert86 site.

 

10 hours ago, parazit said:

Is it possible to only run a post read instead of performing a complete pre clear? I just want to make sure the disk is fine, so i can investigate where the error is coming fro

It is a long time since I ran a pre-clear  I think that you are forced to run the write phase but have options to skip the pre-read and/or post-read.

[parazit]

thanks for your instruction. I ran a memtest and everything was fine without any errors.

 

Just started the pre-clear once again and the log doesn't show any errors so far. Hope it will stay like this  

[parazit]

Unfortunately i still had errors during the post-read. Zeroing the disk was fine.

"4168 sectors are pending re-allocation..." does that mean i should RMA the disk eventhough the SMART Test was good?

WDC_WD6003FFBX-68MU3N0_V9G95D3R-20211108-1309 (sdh).txt

[JonathanM]

7 minutes ago, parazit said:

does that mean i should RMA the disk eventhough the SMART Test was good?

I'd run a long smart test again, followed by another preclear cycle. Is there any possible way to connect directly instead of USB?

[parazit]

I will run another long smart to see the result.

 

All SATA-Ports are occupied but i can disconnect one of the data drives, run the tests on the new HDD and reconnect the drive again.

That should normally work right?

 

EDIT: Just connected the HDD directly and ran the extended SMART test. Unfortunately the test failed. I am going to RMA the disk. Thank you everybody who tried to help me in this case

 

Edited November 8, 2021 by parazit

[itimpi]

48 minutes ago, parazit said:

"4168 sectors are pending re-allocation...

On a new disk I would expect this value to be 0.   Even on an old disk a number this large would be a cause for concern.

[parazit]

Yes this value is 0 on all my disks thats why i don't even want to invest more to time for any other test. The result clearly shows that there is something wrong with the disk eventhough it's brand new.