4 new HDD all fail smart test

[tbvink]

Hi,

I have just built my NAS, trueNAS core 13, and have 4 new Seagate X18 18TB drives in a single pool. I have manually started the long smart test from the disks dialog. Tried twice. All 4 drives come back FAIL with only two lines in the report.

Lifetime: 10​

Error:N/A​

Doubting that the drives are really bad, I am looking for something that I may be doing wrong, or perhaps something in TN that is not compatible with these drives smart tests. Proper smart testing is pretty important so I want to get to the bottom of this.

Thanks,
Tony

 

[ChrisRJ]

Please provide information about your hardware. For the right level of detail, the forum rules (in red at the top middle of the screen) provide pretty good guidance.

 

[tbvink]

Hi,

This should tell you everything...
Version: TrueNAS-13.0-U2

I have just updated the BIOS to E7D25IMS.A30 which came out in March of this year.

Tony

 

[MrGuvernment]

Can you connect the drives to a windows system or other system and run tools against them to confirm if they are infact failing.

This is one issue with buying all your drives from the same place and the same time, you can get a bad batch. Or if a rebuild happens because one drive died, chances are the other drives could also die shortly after.

 

[AlexGG]

Some SMART attribute dumps will be helpful, too. A bum batch of four drives is unpleasant but it does happen. The largest I have ever seen was five out of six.

 

[tbvink]

Can you connect the drives to a windows system or other system and run tools against them to confirm if they are infact failing.

This is one issue with buying all your drives from the same place and the same time, you can get a bad batch. Or if a rebuild happens because one drive died, chances are the other drives could also die shortly after.

Hi,

I did install Windows 11 and the drives pass. However it took over 24 hours to complete. I fear that I may have done something to force the test to terminate prematurely when running in TrueNAS. I am new to TrueNAS and have moved on from that issue.

My thinking is that I would run the quick test nightly. I believe that this would show any errors that were detected in the prior day's normal operation since SMART is enabled. I don't plan on running the 24+ hour test unless something is reported in the quick test.

If you have any insight into the validity of this plan I would appreciate hearing that.

Thanks,
Tony

 

[tbvink]

I did keep the results of the Seagate test that I ran on windows. I have attached a file with screenshots. The "read" and "seek" error rates are presumably normal since the tests pass. I googled these and found statements that these are "pre error correction" and are normal on these newer high density drives.

(taken with a grain of salt... and the size of these integers look like an overflow of a data type.)

Tony

 

[garm]

Please post the printout of smartctl -a <device> (substitute <device> with the actual device from /dev) in code brackets