Hello Everyone!

 

Discovered my first parity drive errors and I'm looking for confirmation on my next steps.

 

Context: Its my first build and have been running for about a month with 5.0b14 and my last parity check was 14 days ago with no errors. Recently transferred some files across my network overnight via Teracopy from my Win 7 machine, and it said bad crc match on only one file (first time that's happened in about 4TB of movies), so I copied it over again, this time with success.  Took a look at unRAID main page and saw 122 errors on the parity drive.

 

I've included the short and long SMART tests (which seem to be OK) as well as the syslog. I think I've posted them correctly, but let me know.

 

After searching and reading through the forum posts, I can't decide which parity check to run (correct or no correct). Since I believe the drive to be OK (I know disks can fail, but I ran successful 3-cycle pre-clears on every drive in my array), I want to run a correcting parity check, but thought I would double-check with the forum first.

 

Third option would be to stop the array, un-assign the parity drive, start the array, stop the array, re-assign the parity, and let it build parity sync, run a parity check, then see if the errors arise again.

 

If someone could give me a nudge in the right direction, I would appreciate it. Most of my relevant info should be in my sig & my apologies for the newb handholding.

 

Thanks  for your help.

syslog-2012-06-01.zip

Parity_Drive_Smart_Test_6_1_12.txt

The parity disk looks ok. The parity issue does not explain the Teracopy errors. Post a SMART report for the data disk.

Thanks for the help dgaschk,

 

I didn't think to check the data drive since the unRAID main menu didn't list any errors. In hindsight, seems like a good idea.

 

I've included the SMART report for the data drive in question (drive 9),and noticed that it does have have some disturbingly high numbers in some categories, but no re-allocation of sectors.  Maybe I should have submitted the pre-clear results for the forum to look at, but I didn't think it was that bad. 

 

I have a replacement drive, but would obviously like to know if I can trust my parity drive at the moment before attempting to rebuild drive 9.

 

I'm still unsure what my next step would be in terms of the correct/no correct parity check.

 

-L

Drive_9_Smart_Test_6_2_12.txt

Personally, I would perform a memory test first. (several passes, overnight if possible)

 

Then, check for disks with re-allocated sectors, or sectors pending re-allocation.

Thanks for the input Joe.

 

I ran a memtest for about 30 hours (something like 28 passes I think) with no errors.  Now, I'll start doing long SMART tests on the rest of the data drives looking for re-allocated sectors or sectors pending re-allocation (see earlier post for the results for drive 9- the one I suspect most) and will post when finished.  Better to figure out this kind of stuff earlier than later, I guess.

 

-L

Finally an update:

 

I ran long (& short) tests on all my data drives (as well as my parity drive) and didn't see any re-allocated sectors or sectors pending re-allocation anywhere.  Some drives had high Raw_Read_Error_Rate errors, but that was it.  I had to reboot the box to run the 30 hour memtest, but since re-starting the array, I haven't done anything else besides run the SMART tests via unMENU. Not surprising, the errors have not reappeared in the parity drive column yet,  so I think I will run a non-correcting parity check and see what happens. I'm not sure what I'll do if any errors do show up, but let me know if anyone else has any other suggestions.

 

I've included a zipped set of the SMART reports (as well as the original syslog) if anyone has the wherewithal to look through them.

 

 

Thanks much,

 

-L

SMART_Reports_plus_syslog.zip