(6.11.5) Week long parity check. Help!

[MaximusFSU]

Hey all.

 

Built my first unraid server about 3 months ago using a mix of some new, and some older (4 years) drives.  Been LOVING it so far.  First parity check after I transferred everything went off without a hitch.  Took a day or two I think.  After that I set it up to run parity checks every 3 months.  But just yesterday I home from a vacation to THIS...

 

The errors were concerning enough, but the time also seemed VERY long for a single 16TB parity drive.  Note: I did not pause it myself... but I have since resumed the operation and it's showing this now:

 

 

I've been searching for troubleshooting best practices for long parity checks / errors and have only been finding a lot of specific use cases.  And not any standardized "start here, then check this..." type of best practices.  So I'm coming to all of you brilliant geniuses for help.  Clearly something is wrong, and I want to sort it before it becomes a bigger problem.  Logs attached.  Let me know if you need any other info.  Server is mostly used for Plex, the arrs, time machine backup for 3 macs, light nextcloud use, and a Blue Iris camera server, running in a Win 10 VM (on cache drive) writing video files to an unassigned surveillance drive.

 

Thanks in advance!!

 

normandy-diagnostics-20230911-1029.zip

[JorgeB]

Post a screenshot of the scheduled parity check settings, it's pausing a lot.

[MaximusFSU]

[JorgeB]

It's only set to run 6 hours per day, is that what you want?

[MaximusFSU]

Ah... It is possible that the problem is that I'm an idiot.  I may have set it up that way early on in my setup and forgot about it. 

 

So I guess the "6 days" is a bit misleading then.  it's really only been running for 6 hours per day.  So in reality it's only been really conducting the parity check for about a day and a half or so (~36 hrs)?  So it's not really taking longer than it should.

 

But the 3 errors should still be concerning, yes?

[JorgeB]

13 hours ago, MaximusFSU said:

But the 3 errors should still be concerning, yes?

Yes, unless there was an unclean shutdown since last check, then it would be normal, run a correcting check, followed by a non correcting, if the 2nd one finds new errors post new diags.

[MaximusFSU]

Thanks Jorge.  Can't recall any unclean shutdowns. Will do and report back. You rock!

[MaximusFSU]

Hey there. Ran a correcting parity check which finished in a normal amount of time but those 3 errors are still there.  New diagnostics attached. Any help or guidance on next steps / best practices is appreciated. 

normandy-diagnostics-20230921-2102.zip

[MrGrey]

On 9/11/2023 at 10:47 AM, MaximusFSU said:

Built my first unraid server about 3 months ago using a mix of some new,

Maybe start again from "scratch" 🙂

 

I don't mean to be mean, but isn't it better to start again after only 3 months? (I did)

 

MrGrey.

 

[itimpi]

2 hours ago, MaximusFSU said:

Hey there. Ran a correcting parity check which finished in a normal amount of time but those 3 errors are still there.  New diagnostics attached. Any help or guidance on next steps / best practices is appreciated. 

normandy-diagnostics-20230921-2102.zip 254.89 kB · 0 downloads

Your first correcting check will report the same number of errors as it lists the number of corrections it made.    Any subsequent checks (preferably non-correcting) should report zero.