Two drives failed at the same time.

[szoemi]

Unraid version 6.9.2

 

Is this just a fluke?

 

disk2 (sdb) and disk3 (sdg) failed at the same time and were taken off-line.

 

This is part of the system log:

 

Apr 12 20:18:52 SPFBACKUP kernel: md: disk2 read error, sector=9454620576
Apr 12 20:18:52 SPFBACKUP kernel: md: disk2 read error, sector=9454620584
Apr 12 20:18:52 SPFBACKUP kernel: md: disk2 read error, sector=9454620592
Apr 12 20:18:52 SPFBACKUP kernel: md: disk2 read error, sector=9454620600
Apr 12 20:18:52 SPFBACKUP kernel: md: disk3 read error, sector=9454620576
Apr 12 20:18:52 SPFBACKUP kernel: md: disk3 read error, sector=9454620584
Apr 12 20:18:52 SPFBACKUP kernel: md: disk3 read error, sector=9454620592
Apr 12 20:18:52 SPFBACKUP kernel: md: disk3 read error, sector=9454620600
Apr 12 20:19:03 SPFBACKUP kernel: md: disk2 write error, sector=9454620576
Apr 12 20:19:03 SPFBACKUP kernel: md: disk3 write error, sector=9454620576
Apr 12 20:19:03 SPFBACKUP kernel: md: disk2 write error, sector=9454620584
Apr 12 20:19:03 SPFBACKUP kernel: md: disk3 write error, sector=9454620584
Apr 12 20:19:03 SPFBACKUP kernel: md: disk2 write error, sector=9454620592
Apr 12 20:19:03 SPFBACKUP kernel: md: disk3 write error, sector=9454620592
Apr 12 20:19:03 SPFBACKUP kernel: md: disk2 write error, sector=9454620600
Apr 12 20:19:03 SPFBACKUP kernel: md: disk3 write error, sector=9454620600

 

Previously, disk3 (sdg) issued a SMART warning: Reallocated sector count = 8
disk2 (sdb) had no SMART errors

 

The system log above happens to show 8 errors, the same number as the reallocated
sector count of disk2.

 

I may be wrong, but I though 8 reallocated sectors is 'nothing'.

What is your expert opinion? Can I safely re-eable these drives?

Their warranty ended 2 years ago, they are both Seagate iron wolfe drives,
bought at the same time.

 

I have 2 parity drives, and 3 data drives.

 

I am in the process of pre-clearing the failed drives. I it goes well, does it mean they are safe to re-use?

 

The drives are all 8 TB.

Edited April 17, 2022 by szoemi
Add more info; add pre-clearing of drives.

[Squid]

You should post the diagnostics

 

Actual drive failures are exceedingly rare.  Usually its because of loose sata cabling

[szoemi]

Thanks for the reply. I attached two diagnostic files, one was just taken, the other earlier, and shows the read/write errors I mentioned in my post, happening on April 12.

spfbackup-diagnostics-20220414-2006.zip spfbackup-diagnostics-20220417-0907.zip

[JorgeB]

Apr 12 08:43:22 SPFBACKUP kernel: ata6.00: disabled
Apr 12 08:43:22 SPFBACKUP kernel: ata6.00: detaching (SCSI 6:0:0:0)
...
Apr 12 08:44:31 SPFBACKUP kernel: ata1.00: disabled
Apr 12 08:44:31 SPFBACKUP kernel: ata1.00: detaching (SCSI 1:0:0:0)

 

Both disks dropped offline in the space of a minute, this is usually a power/connection problem, start by checking if they share anything in common, like a splitter.