29254 errors after a scheduled Parity Check

[Artanis]

Hello everyone,

I'm a new Unraid user (just bought the Plus version a couple of weeks ago). A few minutes ago a scheduled Parity Check just completed.

Turns out there were 29254 sync errors it automatically corrected.

 

 

The previous two parity checks were without any error and I didn't change anything (the PC done on March 30th was after I added a 2TB disk and showed zero errors).

Afaik there was no unclean shutdown of the UnRaid system (it shuts itself down after a few minutes of inactivity at midnight via the Dynamix S3 Sleep plugin and I usually see no errors in the logs although they get wiped with each login).

I'm a bit concerned, although the system itself seems to be working properly. I'm attaching the diagnostic. Can you please let me know what should I do and if I should be concerned?

There are no discernible errors in the disks:

 

And they all show up as Healthy in the SMART checks.

Thanks for your help!

artaunraid-diagnostics-20210412-1657.zip

Edited April 12, 2021 by Artanis
Typos

[JorgeB]

Without rebooting run a non correcting check, if more errors are found post new diags and there's a problem for sure, if not most likely the result of an unclean shutdown since last check.

[Artanis]

Thanks. I'll start the parity check without correcting checks and I'll report back (probably tomorrow since it will take 8 hours).

[itimpi]

38 minutes ago, Artanis said:

A few minutes ago a scheduled Parity Check just completed.

Turns out there were 29254 sync errors it automatically corrected.

It is normally recommended that scheduled parity checks are set up as non-correcting.    That rationale is that if you have a drive mis-behaving and producing read errors you will not end up corrupting parity inadvertently.   Ideally you only want to run a correcting check when you think there might be some errors to correct and you know why they happened.

[Artanis]

Thanks, I'll keep this in mind. I thought it was OK as a setting (fixing right away) similarly to a chkdsk, but I see now how this might not be desirable.

Speaking of errors, if it turns out to be "non-graceful-shutdown" related (which might be then caused by the Sleep plugin), should I set it up with a script to stop running docker containers? Maybe that's the issue... I assumed that the shutdown command (the same you can issue manually) toke care of gracefully stopping everything (Array included), but maybe it's not so...

p.s. 25% parity check, 0 errors for now.

[Artanis]

Reporting as promised; the parity check was completed with 0 errors.

 

 

So can I stop worrying, was it due to unclean shutdown and I should double check the Sleep plugin? Maybe I really do need to manually stop the docker containers with some script?

Thanks!