Errors on brand new 8TB WD White Label drives

[Random.Name]

Well, since i needed to upgrade my storage and the local hardware store had a good deal on WD my Books (160€) i bought one and replaced my parity drive...

During the build process the array was taken offline due to some errors (around 300) on the new drive. I tried a couple of things to fix the problem (taking out my VM GPU and my USB3.0 PCIe card) and ultimately re-wired all my drives. Also the smart report (https://pastebin.com/YwyWg7hx) looked fine to me and i spun the array up again.

Everything re-build fine and seems to be running smooth now.

So after that i bought a second 8TB my Book and added that one to the array...pre-cleared it and all. Again the drive failed after a few hours, with around 1000 Errors.

Again, the smart logs look fine to me (https://pastebin.com/J0S7hZLY)

My personal guess is, it's the SATA data cables. i have a few old ones, without clips and maybe they don't work right. I ordered some new ones and will try to start the array again.

 

My most pressing questions are, do i miss something in the smart logs? I am still within a timeframe to return the drives and would do if they are a risk.

Ans also might there be an other problem that i could look into? Second idea was, that my power supply (550W) might be to weak for 9 drives, PICe USB3 and a Nvidia 750... anyway, here is my hardware profile https://pastebin.com/acuiwdDi And if you need further information to help me please ask and i will provide them as soon as i can.

 

 

[SSD]

1 hour ago, Random.Name said:

Well, since i needed to upgrade my storage and the local hardware store had a good deal on WD my Books (160€) i bought one and replaced my parity drive...

During the build process the array was taken offline due to some errors (around 300) on the new drive. I tried a couple of things to fix the problem (taking out my VM GPU and my USB3.0 PCIe card) and ultimately re-wired all my drives. Also the smart report (https://pastebin.com/YwyWg7hx) looked fine to me and i spun the array up again.

Everything re-build fine and seems to be running smooth now.

So after that i bought a second 8TB my Book and added that one to the array...pre-cleared it and all. Again the drive failed after a few hours, with around 1000 Errors.

Again, the smart logs look fine to me (https://pastebin.com/J0S7hZLY)

My personal guess is, it's the SATA data cables. i have a few old ones, without clips and maybe they don't work right. I ordered some new ones and will try to start the array again.

 

My most pressing questions are, do i miss something in the smart logs? I am still within a timeframe to return the drives and would do if they are a risk.

Ans also might there be an other problem that i could look into? Second idea was, that my power supply (550W) might be to weak for 9 drives, PICe USB3 and a Nvidia 750... anyway, here is my hardware profile https://pastebin.com/acuiwdDi And if you need further information to help me please ask and i will provide them as soon as i can.

 

 

 

Smart report for drive looks fine. Looks like you ran a SMART test that completed successfully.

 

Where did you see 300 errors? Suggest posting diagnostics file.

 

Are you running disk as a USB? Or did you shuck the drive? Assuming you shucked it.

 

Cabling is the most common problem with installing new drives.

[Random.Name]

2 hours ago, SSD said:

 

Smart report for drive looks fine. Looks like you ran a SMART test that completed successfully.

 

Where did you see 300 errors? Suggest posting diagnostics file.

 

Are you running disk as a USB? Or did you shuck the drive? Assuming you shucked it.

 

Cabling is the most common problem with installing new drives.

Oh, sorry missed some parts there.
EDIT: Yes, i ran a short and an extended smart test.
I saw the errors in the unRAID Main Tab, next to read and write.  
Diagnostics are here now , should have thought about them in the first post.

 

Yes i shucked the drives, and i used a molex adapter to power them because WD does some trickery with the 3,3V lane.

The Molex adapter is brand new, too.

 

Yeah, i really hope the new cables will fix that issue...very annoying

Thanks for the help

diagnostics-20180714-2300.zip

Edited July 14, 2018 by Random.Name

[JorgeB]

Most the likely the problem is the Marvell controller, these are known to drop disks regularly for many users.

 

P.S. disk2 needs a new SATA cable.

 

 

[Random.Name]

On 7/15/2018 at 12:04 PM, johnnie.black said:

Most the likely the problem is the Marvell controller, these are known to drop disks regularly for many users.

  

 P.S. disk2 needs a new SATA cable.

 

 

 

Not quite sure here, but the Marvell controller is the PCIe SATA card? 

That would be my best guess, now too. The Sata Card has 2x2 internal ports and it seem like the left one of those ports drops the HDDs.

Have everything running on the right two ports now and everything seems fine.

Any suggestions for a replacement SATA/ HBA card?

 

will look for new cables, soon

 

[JorgeB]

1 hour ago, Random.Name said:

but the Marvell controller is the PCIe SATA card? 

Yes.

 

1 hour ago, Random.Name said:

Any suggestions for a replacement SATA/ HBA card?

Any LSI with a SAS2008/2308/3008 chipset in IT mode, e.g., 9201-8i, 9211-8i, 9207-8i, 9300-8i, etc and clones, like the Dell H200/H310 and IBM M1015, these latter ones need to be crossflashed.

[Random.Name]

8 hours ago, johnnie.black said:

Yes.

 

Any LSI with a SAS2008/2308/3008 chipset in IT mode, e.g., 9201-8i, 9211-8i, 9207-8i, 9300-8i, etc and clones, like the Dell H200/H310 and IBM M1015, these latter ones need to be crossflashed.

Thanks for the help, and now that i have seen your exact answer for HBA Cards in multiple threats i feel stupid for asking

Just ordered a used Dell H310, and hopefully this will solve my Problems.

[pwm]

9 minutes ago, Random.Name said:

Thanks for the help, and now that i have seen your exact answer for HBA Cards in multiple threats i feel stupid for asking

Just ordered a used Dell H310, and hopefully this will solve my Problems.

 

Always better to ask once to much than once to little.

[Random.Name]

On 7/17/2018 at 10:19 AM, johnnie.black said:

Yes.

 

Any LSI with a SAS2008/2308/3008 chipset in IT mode, e.g., 9201-8i, 9211-8i, 9207-8i, 9300-8i, etc and clones, like the Dell H200/H310 and IBM M1015, these latter ones need to be crossflashed.

 

Ok, here I am again

the Dell h310 just arrived and I swapped it in my server.

all drives seem to be recognized, but they got new names.

so unRAID now does not recognize one of my data and my cache drives.

is this normal? I guess since it is only one data drive I can rebuild it. But since it’s the 8TB drive it will take ages.

 

i think the hba is already in it mode since all drives are recognized. But maybe I am wrong? 

 

Anny suggestions?

[JorgeB]

4 minutes ago, Random.Name said:

all drives seem to be recognized, but they got new names.

Post your diagnostics but if the names are different it's likely in IR mode, flash or crossflash to LSI IT mode

[Random.Name]

2 minutes ago, johnnie.black said:

Post your diagnostics but if the names are different it's likely in IR mode, flash or crossflash to LSI IT mode

Thanks a lot

Also during Boot i get a message

 

Quote

All of the disks from your previous configuration are gone. If this is an unexpected message, then please power off your system an check your cables to ensure all disks are present. Press any key to continue, or 'c' to load the configuration utility.

I did not press 'c', but continued the boot process.

daxserver-diagnostics-20180721-1335.zip

[JorgeB]

It's in IR mode.

[Random.Name]

9 hours ago, johnnie.black said:

It's in IR mode.

ok....just a question. Is there any problem with IR mode if i can see all drives?

For the last 9 hours i tried to flash the card but with 0 success.

Problem is/ seems to be: I only have a company locked macBook. So i can not really install anything useful on here to create a FreeDOS usb stick.
The other laptop that i have is an ancient eeepc netbook, that only runs ubuntu 17.04, this version is EOL and won't let me install anything useful either. I tried installing Debian on there but every time i try to run UNetbootin i get a warning saying i can only run the program as root. If i try to start the program as su from the Terminal i get some errors about BadAccess and the UNetbootin window stays grey.

 

Or well... it seems like i get the FreeDOS stick to work. I can boot from it and i get the Unetbootin boot prompt, where i can only select default.

After that i get a freedos boot loader, asking me if i want to install freedos to a disk, or start the live cd (multiple options here)

No matter what i choose i get some error about no cd drive beeing found and i get a command prompt "A:" 

only instaed of all the files i copied onto the stick, there are only two folders, "Drivers" and "freedos"
and i dont seem to be able to do anything

One reason for having a unRAID server is having Ubuntu and Windows VMs to do stuff i can't do on my macbook
 

Edited July 21, 2018 by Random.Name

[JorgeB]

12 hours ago, Random.Name said:

ok....just a question. Is there any problem with IR mode if i can see all drives?

IT mode uses a different driver, more commonly used and likely to have fewer issues, also the disks identifiers will be changed if you change the controller, requiring a new config.

[Random.Name]

1 hour ago, johnnie.black said:

IT mode uses a different driver, more commonly used and likely to have fewer issues, also the disks identifiers will be changed if you change the controller, requiring a new config.

Yayee...after countless hours i finally got it

Thanks for all the initial help. and help me god i did not expect to really need a windows computer to create the DOS Stick... but after that it worked...