Parity errors fixed but no errors showing up in gui?

[dnoyeb]

Needing a little help; got an email today about 3645 errors; ran it again since I realized my monthly check wasn't setup to auto correct.  Same number, shows that it fixed it.

 

This system has been running for almost 10 years at this point; migrated to new mobo/cpu years ago, but there are drives in there since 2012.  

 

Anyways, not sure what to make of it, since in the gui it doesn't show any drive specific errors...  I did see in my log that the cache drive has some sort of metadata store error and recommends repairing; but don't see how / why that would result in parity errors.  

 

Can someone take a peak at the diag file and advise on how I can tell if there is a drive needing to be replaced?  I've got 2 extra 12tb sitting in there just waiting to be swapped into the array (precleared already)...

 

Thanks!

unraid-diagnostics-20210517-1039.zip

[JorgeB]

May  4 19:20:31 unRAID emhttpd: unclean shutdown detected

A few sync errors are normal after this.

[dnoyeb]

What's strange is that day, the system was just completely locked up.  first time in the 10 year's I've had the system that it has happened....  That and it did a parity check on startup and didn't find anything.

Edited May 17, 2021 by dnoyeb

[JorgeB]

19 minutes ago, dnoyeb said:

That and it did a parity check on startup and didn't find anything.

Yep, that rules out the unclean shutdown and makes the sync errors much more important, since RAM is ECC and the second check found the same errors that's unlikely the reason, first suspect would be a disk.

 

 

 

[JorgeB]

Forgot to mention, you need to run a filesystem check on the cache device.

[dnoyeb]

41 minutes ago, JorgeB said:

Forgot to mention, you need to run a filesystem check on the cache device.

Guess I'll start with doing the file system check on that device...  Other than that; at a bit of a loss since the pool's disks don't show any errors or smart flags.  

 

This thing has been such a rock for so many years!  

 

[JorgeB]

7 minutes ago, dnoyeb said:

Other than that; at a bit of a loss since the pool's disks don't show any errors or smart flags.  

That's not unheard of if it is a disk, though unless it keeps happening practically impossible to find the culprit.

[dnoyeb]

ah, actually disk 7 has reallocated sectors.  Sounds like candidate to replace...  Now to figure out how to upgrade parity to 12tb (from 8tb) and not lose my parity; then use the 8tb parity drive to become disk 7....

 

 

also strange is just about every newer drive in my system show signs of "Raw read error rate" and "Hardware ECC recovered"... not really sure what to make of that (haven't really looked at that before to know if it's recently increased or not).    

 

Only that one disk 7 has reallocated sectors however.  

 

Edited May 17, 2021 by dnoyeb

[JorgeB]

8 minutes ago, dnoyeb said:

also strange is just about every newer drive in my system show signs of "Raw read error rate" and "Hardware ECC recovered"... not really sure what to make of that (haven't really looked at that before to know if it's recently increased or not).    

Those are normal with Seagate drives.

[dnoyeb]

9 minutes ago, JorgeB said:

Those are normal with Seagate drives.

thanks man.  ok so then really there is only one drive that has any "real" errors to it, which are reallocated sectors.  will keep an eye on things going forward and see if anything increases; pulled a diag from 2019 and that same drive had the same 8 reallocated sectors...  so probably not leading to the issue.  

 

Thanks again so much for your help.  Time will tell.

Edited May 17, 2021 by dnoyeb