Disk has read errors. Smart is fine, parity check ok.

[Howboys]

Got a notification 2 days ago that disk 5 has 78 read errors. But SMART shows no errors and parity check succeeded finding 0 errors.

 

Any help please?

 

Edited November 3, 2022 by Howboys

[trurl]

Run an extended SMART self-test on disk5.

 

What are all those SSH key exchange errors from WAN IP in your syslog about?

[Howboys]

Okay starting a smart test now.

 

Those errors I think are from tailscale, but I'm not sure exactly why. They're annoying but I've learned to ignore them.

[Howboys]

10 hours ago, trurl said:

Run an extended SMART self-test on disk5.

 

What are all those SSH key exchange errors from WAN IP in your syslog about?

Completed without error.

tower-smart-20221031-1628.zip

[JorgeB]

See here:

 

[Howboys]

26 minutes ago, JorgeB said:

See here:

 

 

Thanks. Before I try that, I'm curious why you think it's the same issue. My drives don't disappear. Just this one has read errors but unRAID doesn't seem to show what exactly the errors are.

 

I'm also not sure how/if to rest the error counter. 

Edited November 1, 2022 by Howboys

[trurl]

2 hours ago, Howboys said:

the error counter

Main - Array Operation - Clear Stats

[JorgeB]

2 hours ago, Howboys said:

I'm curious why you think it's the same issue

Because you are using an LSI HBA with a known affected Ironwolf disk model and the errors happened just after spin up.

[Howboys]

1 hour ago, JorgeB said:

 the errors happened just after spin up.

 

Oh interesting. For me knowledge, where in the logs do you see that?

 

Separately, is there any indication that the drive is failing or data on drive is corrupted/at risk? Are the reported errors (78) just a red herring?

Edited November 1, 2022 by Howboys

[JorgeB]

1 hour ago, Howboys said:

where in the logs do you see that?

The syslog.

 

1 hour ago, Howboys said:

Separately, is there any indication that the drive is failing or data on drive is corrupted/at risk? Are the reported errors (78) just a red herring?

No, just looks like the linked issue.

[Howboys]

Okay I've disabled EPC and Low Current Profile on the 3 drives matching that model. After a reboot, disk error count is back to 0. Hopefully this doesn't happen again. Thanks!