Hi,

 

I have been running unRAID fine for almost a year after without any problems.

1. June I first found an error when running my monthly Parity check. The same error came month after month (I was hoping it would get corrected by it selves)

 

Where do I begin if I what to fix this error? I have attached my diagnostics file, hope some of you can help me with this.

 

Event: unRAID Parity check

Subject: Notice [NAS] - Parity check finished (1 errors)

Description: Duration: 7 hours, 4 minutes, 18 seconds. Average speed: 117.9 MB/s

Importance: warning

 

 

nas-diagnostics-20170801-2314.zip

12 minutes ago, ahoo said:

(I was hoping it would get corrected by it selves)

 

It probably just did, the most logical explanation would be that the first error was the result of a bit flip (likely from a memory error since you're not using ECC) and parity was incorrectly updated, and this month it was corrected, if next month there are no errors you are fine, consider a board with ECC if/when you upgrade your current server.

The error has been the every month since June. So it looks like its not going away by it selves. Any ide how to correct error?

 

Edited August 2, 20178 yr by ahoo

Do you have parity check set to correct parity errors?

 

Settings - Scheduler - Parity Check - Write corrections to parity disk

Its set it to: "Write corrections to parity"

Edited August 2, 20178 yr by ahoo

In my main window it does not show any read/write Errors on any of the disks. Would an parity error show here? Sorry for the stupid questions.

 

Sync errors are different from read/writes errors, only those will appear on the array errors column.

1 hour ago, ahoo said:

The error has been the every month since June.

 

Well there goes my theory then, according to the log it was always the same sector, at least for the last 3 times, so my nest guess would be a disk, start by doing an extended SMART test on each disk, but doubt it would identify this kind of problem, if it doesn't your best bet may be to replace one disk at a time, luckily you only have 3.

How do I run a SMART test? And is there not a way to tell what disk this error is on?

 

Has anyone tryed SpinRite (grc.com) on a disk witch errors?

Edited August 2, 20178 yr by ahoo

Just now, ahoo said:

How do I run a SMART test?

 

Click on a disk and scroll down.

 

1 minute ago, ahoo said:

And is there not a way to tell what disk this error is on?

 

No

I ran the SMART extended self-test: Completed without error

 

So what is my next step? 

Edited August 6, 20178 yr by ahoo

Bump

On 02/08/2017 at 8:54 PM, johnnie.black said:

start by doing an extended SMART test on each disk, but doubt it would identify this kind of problem, if it doesn't your best bet may be to replace one disk at a time, luckily you only have 3.