Parity checks reporting the same number of errors with each run

[dotsonic]

I built a new Unraid sever earlier this year and dropped in 3x10TB, and 2x6TB drives, with 1 of the 10TB drives acting as parity.  The disks all formatted and worked properly.

 

About a month later I experienced a power outage and re-ran parity after it came back online.  The parity check reported 2017 errors but reported all corrected.  My first scheduled/automated parity check ran earlier this week (8/1) and completed, again reporting 2017 errors.

 

 

Is there a step that needs to be performed to clear the errors?  Or do I have another issue (ex. bad drive)?  I've run the fast SMART tests against all drives and they all came back clean.

[trurl]

Your description doesn't really say if you ever ran a correcting parity check.

 

22 minutes ago, dotsonic said:

re-ran parity after it came back online

When you have an unclean shutdown, it automatically starts a noncorrecting parity check. Is that the "re-ran" you are talking about?

 

24 minutes ago, dotsonic said:

parity check reported 2017 errors but reported all corrected

Where did you see that it said they were all corrected?

 

25 minutes ago, dotsonic said:

scheduled/automated parity check

Hopefully you have that set as noncorrecting also.

 

Do you have diagnostics or syslog from any of these parity checks?

[dotsonic]

34 minutes ago, trurl said:

Your description doesn't really say if you ever ran a correcting parity check.

When you have an unclean shutdown, it automatically starts a noncorrecting parity check. Is that the "re-ran" you are talking about?

 

Thanks for pointing this out - I wasn't aware parity check didn't automatically correct errors.  I now see that there's a checkbox next to the 'Check' button to 'write corrections to parity'

 

34 minutes ago, trurl said:

Hopefully you have that set as noncorrecting also.

 

The scheduled task is set to 'Write corrections to parity disk':

 

 

At this point do you recommend that I start a manual parity check with the 'write corrections to parity option' enabled?

 

I'm in the process of creating a diagnostics file and will post once complete.

 

Appreciate the help!

 

 

 

[itimpi]

56 minutes ago, dotsonic said:

The scheduled task is set to 'Write corrections to parity disk'

It is normally recommended that the scheduled checks do NOT automatically correct errors.   The reason is that you if you have a drive misbehaving and you are not aware of it you do not want it corrupting parity.   If the scheduled check reports anything other than zero errors then it is time to investigate why.   Ideally you should always run correcting checks manually after you have fixed any hardware issues.