GUI saying one thing... zpool says another.

[Robin Ball]

Hi folks,

I have an odd one for you. My GUI is saying there is a problem with one of my disks... zpool does not. Anyone come across this before?

Cheers, Rob


ZPOOL STATUS OUTPUT:

pool: ZFS16

state: ONLINE

scan: scrub repaired 0 in 20h1m with 0 errors on Sun Nov 12 20:01:05 2017

config:


NAME STATE READ WRITE CKSUM

ZFS16 ONLINE 0 0 0

raidz2-0 ONLINE 0 0 0

gptid/4f23162a-eb4e-11e4-afc8-e03f490de638 ONLINE 0 0 0

gptid/500a51ef-eb4e-11e4-afc8-e03f490de638 ONLINE 0 0 0

gptid/5070e197-eb4e-11e4-afc8-e03f490de638 ONLINE 0 0 0

gptid/50cea95c-eb4e-11e4-afc8-e03f490de638 ONLINE 0 0 0


errors: No known data errors



GUI ERROR:

• CRITICAL: Nov. 25, 2017, 8:20 p.m. - Device: /dev/ada1, 1 Currently unreadable (pending) sectors
• CRITICAL: Nov. 25, 2017, 8:50 p.m. - Device: /dev/ada1, Self-Test Log error count increased from 0 to 1

 

[garm]

That is a S.M.A.R.T error and not a ZFS error.
I would replace the drive and then run a long test etc on a different machine

 

[rs225]

That is normal. The error isn't going to show up with ZFS; it is the SMART monitoring letting you know that there is a marginal sector that the drive will attempt to relocate the next time it is written. You can ignore it, attempt to force relocate it, or replace the drive. I would recommend doing nothing, but keep an eye for any further changes in error count.

 

[Robin Ball]

Thanks a million lads. I'll keep a drive on standby I think.

 

[Jailer]

Is the drive under warranty? If so RMA it.

 

[Robin Ball]

Is the drive under warranty? If so RMA it.

Out of warranty...

 

[Jailer]

Keep an eye on it then and if the errors increase replace it.