12 minutes ago, Mason736 said:

and SMART is failing

If SMART is filing it's another problem, you should post the complete diags.

So I woke up this morning, and two of the ST8000VN04 drives had dropped from the array with read errors.  They are both brand new drives from Seagate.  
 

I'm convinced is the issue with the ERC and low voltage spin up, so I followed the above instructions.  We will see if they, or any of the other VN04 drives drop out of the array again.

Further update....I followed the steps and disabled EPC and low voltage spin up.  However, BOTH ST8000NV004 drives are still dropping from the array with read errors.

 

Post diagnostics

Here are the diagnostics. The array is currently being rebuilt.  I can post additional ones in a couple hours once it has completed.

archie-diagnostics-20210413-0857.zip

1 hour ago, Mason736 said:

Here are the diagnostics

LSISAS2008: FWVersion(20.00.02.00)

You should upgrade firmware to latest, all p20 releases except latest one (20.00.07.00) have known issues.

LSI 92118i HBA has been updated to 20.00.07.00.  See updated diagnostics.

 

 

archie-diagnostics-20210413-1259.zip

26 minutes ago, Mason736 said:

LSI 92118i HBA has been updated to 20.00.07.00.

Let us know if it helps.

Update: the updated firmware on the 9211-8i definitely helped.  However, i did receive a couple new ST8000NV04 drives, and just stuck them in the array, without disabling EPC or Low Voltage Spin Up.  They dropped from the array almost instantly. Once I used the seagate tools and disabled the features, they have been just fine.  I'm looking forward to whatever the fix may be for the seagate drives.

Update again:  the ST8000V04 drives are not continuing to drop out of the array. I'm unsure how to proceed at this point.

4 minutes ago, Mason736 said:

not continuing to drop

So they are working?

As of yesterday, all of the ST8000NV04 drives are dropping out of the array.  The ST8000NV0022's are working perfect.

Here is a copy of the latest smart extended test.  Keep in mind, this drive is brand new.  I'm not sure how to interpret this.

archie-smart-20210423-0922.zip

Disk has pending sectors and the SMART test failed, so it needs to be replaced. 

awesome.  i'll start the RMA process again, for a brand new drive.  Starting to really not like Seagate.

 

Separately, on a good note, I changed the spin-up of all of the ST8000VN004 drives to never spin down, and none of them have dropped from the array.    The drop/error has something to do with the spin-up, but I'm not technically advanced enough to understand the mechanics behind it.

I need the advice what to do with this drive.

Model/ID: ST6000DM003-2CY186_ZR10CJF8. I don't remember when exactly I got it, but should be no older than 2 years, used very moderately.

 

This was a data drive in a newly created array (4x 6TB, with 1 parity), all attached through the "Serial Attached SCSI controller: Broadcom / LSI SAS2308 PCI-Express Fusion-MPT SAS-2 (rev 05)." (as listed in devices) During the initial parity build it reported a huge number of errors. I assumed it was due to a loose SATA cable, checked/tightened all cables carefully. Upon trying to write to this drive's share, Unraid disabled it. I recreated the array, checked cables again - still the same result, the drive is being disabled after any write attempt. Quick SMART test gives no errors. SMART is attached.

 

I removed the drive from the array for now. What should I do next? Should I just trash it or what?

ST6000DM003-2CY186_ZR10CJF8-20210427-1218.txt

SMART looks fine, but you can run an extended SMART test to confirm it's good.

2 hours ago, temkins said:

recreated the array, checked cables again - still the same result, the drive is being disabled after any write attempt

You should post complete Diagnostics before rebooting when this happens so we can see what is causing this. Connection problems are more likely than disk problems.

8 hours ago, trurl said:

You should post complete Diagnostics before rebooting when this happens so we can see what is causing this. Connection problems are more likely than disk problems.

 

Well, I just tried. I created a share which should use the questionable drive exclusively and tried to write a file to it, Right after that i got Unraid crashed - here is a screenshot of the kernel panic. I have no idea what's happening, the unraid setup is quite mint, I had just 2 dockers running at the moment (Heimdall and Cloudflaredydns).

 

 

Post diagnostics. We don't know anything about your system except that one SMART report

Brand: Seagate IronWolf

Model: ST4000VN008-2DR166

Approximate age at failure: 0.5 years

Drive in warranty: Yes

 

Drive just dropped offline without signs and unable to redrive smart report then. This report was generated after i had the disk removed, rebuild my array with another drive and reconnected the bad drive again into the server as an unassigned drive in order to get the report. 

 

Is my drive dead? Need some advise what to do with the drive. Can i continue to use it or should i just replace it?

unraidhserver-smart-20211222-0941.zip

@AlexKoh, there are only short tests ; the Extended SMART tests are 'aborded by host'. You should disable spin down for the drive and try to run the Extended test again.

9 hours ago, ChatNoir said:

@AlexKoh, there are only short tests ; the Extended SMART tests are 'aborded by host'. You should disable spin down for the drive and try to run the Extended test again.

 

@ChatNoir No wonder my extended test keeps failing. Let me go find out how to disable spin down thank you very much. Will share my logs again once i have acquired it 

1 hour ago, AlexKoh said:

Let me go find out how to disable spin down

On a specific drive, you can set spin down delay from default to never.

8 hours ago, ChatNoir said:

On a specific drive, you can set spin down delay from default to never.

Thanks managed to get it done.

 

Here are the extended test logs. Seeking help if possible. Thanks in advance.

unraidhserver-smart-20211223-1022.zip