Pool state degraded

[Victor87]

Hello all

I need some help to understand why one of my Crucial MX500 2Tb is starting to fail. Have only few months and around 6Tb writing on it.

Attached are also the smart results

16.7 KB file on MEGA

mega.nz

7.1 KB file on MEGA

mega.nz

Thank you

 

[NugentS]

Take it out, put it in a windows machine and run crucial tests on it - see what happens

 

[Victor87]

Take it out, put it in a windows machine and run crucial tests on it - see what happens

Thanks for answering.
Is heavy... My server is at 2000 km far away.
But on smart result the ssd appears ok or do you see some fails? Possible to be some data corruption?
The data on this server is not so important. If it fails doesn't matter. My question is why is creating this failure because I use same ssd's on other server where are important stuff.

 

[Victor87]

Take it out, put it in a windows machine and run crucial tests on it - see what happens

On the ends looks like truenas is reporting completely wrong the failure of my ssd.
What should I do next to fix the degraded pool error?

 

[artlessknave]

sometimes drives just fail. this is why RAID(z) exists, to mitigate the known fact that drives just fail.

if ZFS reports a drive as returning bad data, smart is irelevant; the drive returns bad data. smart failures are merely one way to predict future drive failures, not the only way to know a drive is failing.
this can be caused by bad cables or other connections, or a bad storage controlller, and you will need to investigate all of these possibilities.

 

[Victor87]

sometimes drives just fail. this is why RAID(z) exists, to mitigate the known fact that drives just fail.

if ZFS reports a drive as returning bad data, smart is irelevant; the drive returns bad data. smart failures are merely one way to predict future drive failures, not the only way to know a drive is failing.
this can be caused by bad cables or other connections, or a bad storage controlller, and you will need to investigate all of these possibilities.

Hi

Everything will fail sooner or later :)
I don't use raidz there because is just an server lab to test and learn stuff on it. Sata cable was replaced today and another scrubbing doesn't fix the error. I will leave it like it is to see what's happens in the future. Everything still works great at the moment.

 

[artlessknave]

a scrub doesnt fix errors, it detects them. zfs will automatically fix any errors it can *if* there is redundancy.

you can zpool clear to clear the errors and see if more occur

 

[Victor87]

a scrub doesnt fix errors, it detects them. zfs will automatically fix any errors it can *if* there is redundancy.

you can zpool clear to clear the errors and see if more occur

Error gone. Should I run new scrub?

Thank you very much for your advice

 

[Daisuke]

Error gone. Should I run new scrub?

Yes, that's the hole idea. If the scrub shows an error after clear, then the disk will be trash soon.

 

[mvipe01]

Weird things can happen to drives, I have had a Samsung drive give errors, act perfectly fine and pass all tests and then fail. You also have some unexpected power loss counts, I would guess that a potential power outage could also cause it to throw an error and drop out.

 

[Victor87]

Weird things can happen to drives, I have had a Samsung drive give errors, act perfectly fine and pass all tests and then fail. You also have some unexpected power loss counts, I would guess that a potential power outage could also cause it to throw an error and drop out.

I run 4 scrub test in the past days. Looks like everything is fine. No errors anymore on that ssd. Now I get only one on checksum for the second ssd. I believe is some file corruption. When I have time I will rebuild the pool again.