[SOLVED] Array health report [FAIL]!?

[2TC]

Hi,

 

I have some issues with my parity drive, or so I think.

My report says the following
 

Notice [NAS] - array health report [FAIL]
Parity - ST8000VN0022-2EL112_ZA16XZXN (sdc) - active 40 C (disk has read errors) [NOK]

Parity is valid

 

When looking into the logs of the disk, I can see the following

 

Sep  1 14:21:45 nas kernel: ata2.00: exception Emask 0x0 SAct 0x7f000000 SErr 0x0 action 0x0
Sep  1 14:21:45 nas kernel: ata2.00: irq_stat 0x40000008
Sep  1 14:21:45 nas kernel: ata2.00: failed command: READ FPDMA QUEUED
Sep  1 14:21:45 nas kernel: ata2.00: cmd 60/40:c0:90:1c:c9/05:00:3b:03:00/40 tag 24 ncq dma 688128 in
Sep  1 14:21:45 nas kernel: ata2.00: status: { DRDY SENSE ERR }
Sep  1 14:21:45 nas kernel: ata2.00: error: { UNC }
Sep  1 14:21:45 nas kernel: ata2.00: configured for UDMA/133
Sep  1 14:21:45 nas kernel: sd 2:0:0:0: [sdc] tag#24 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=3s
Sep  1 14:21:45 nas kernel: sd 2:0:0:0: [sdc] tag#24 Sense Key : 0x3 [current] 
Sep  1 14:21:45 nas kernel: sd 2:0:0:0: [sdc] tag#24 ASC=0x11 ASCQ=0x4 
Sep  1 14:21:45 nas kernel: sd 2:0:0:0: [sdc] tag#24 CDB: opcode=0x88 88 00 00 00 00 03 3b c9 1c 90 00 00 05 40 00 00
Sep  1 14:21:45 nas kernel: I/O error, dev sdc, sector 13887937680 op 0x0:(READ) flags 0x0 phys_seg 168 prio class 0
Sep  1 14:21:45 nas kernel: ata2: EH complete

 

On the Main page, I can see that the Parity have "Errors: 168", while when I check the SMART status for the drive it reads as follows

Last SMART test result: Completed without error

 

And the SMART overall-health: says

SMART overall-health: Passed

 

Attached is the SMART test result. 

Let me know if any you need any other details.

 

The ask is, should I replace my drive, or is this another issue?

ST8000VN0022-2EL112_ZA16XZXN-20230911-1349.txt

Edited September 14, 2023 by 2TC
Marking as solved

[trurl]

SMART looks OK and it passed an extended self-test recently.

 

14 minutes ago, 2TC said:

is this another issue?

Attach diagnostics to your NEXT post in this thread.

[2TC]

Please see attached diagnostics

nas-diagnostics-20230911-1804.zip

[JorgeB]

It's logged as a disk problem, but SMART test passed, reboot to clear the errors, you can also swap cables with a different disk to rule that out and keep monitoring.

[2TC]

17 hours ago, JorgeB said:

It's logged as a disk problem, but SMART test passed, reboot to clear the errors, you can also swap cables with a different disk to rule that out and keep monitoring.

Thank you! I will try this and see if it sorts the issue. I have already ordered a spare disk in case it would fail! Will update with result!

 

Edit: After a reboot, no change of cables, there is no report of errors. will run a parity check to be certain, but seems to have sorted the issue. 

Thank you!

Edited September 12, 2023 by 2TC
added more info

[trurl]

3 hours ago, 2TC said:

After a reboot, ..., there is no report of errors

That will always happen since the Errors column counters reset when you reboot. You can also see that the Read and Write columns reset. You can also reset them in Main - Array Operation - Clear Stats.

[2TC]

1 hour ago, trurl said:

That will always happen since the Errors column counters reset when you reboot. You can also see that the Read and Write columns reset. You can also reset them in Main - Array Operation - Clear Stats.

 

That is true, I should see them more as statistics overtime than anything else. The error in the disk log is also gone, but I assume that correlate to the error counter changing from 0 to another number.

I am currently doing a parity check to see if there will be any errors during that, once it is done and all is good I will redo the smart test to see if that can see anything. 

[trurl]

8 hours ago, 2TC said:

The error in the disk log is also gone

What exactly do you mean by "disk log"? If you mean syslog entries about that disk, syslog also resets on reboot.

[2TC]

8 hours ago, trurl said:

What exactly do you mean by "disk log"? If you mean syslog entries about that disk, syslog also resets on reboot.

 

Ah, I assume that is what I mean. When I click the icon on the disk which ways "disk log information". If that is tied to syslog, then yes. Surely syslog gets cleared every reboot, but if it was a read error I would assume it should come up again if there was an issue with the disk.