Can someone help me determine what is causing these errors during the check?  I did have 3 unclean shutdowns about a week ago.

 

If it isn't the unclean shutdowns, what am i looking for in the logs?

tower-diagnostics-20161025-0733.zip

Most likely they are from the unclean shutdowns, if you get some without one then there might be a problem.

Thanks Johnnie!

You should do another check to make sure everything is OK now.

Ok, just started check and will post results. Last one took a little over a day.

 

edit - It started out immediately with read errors, see the following from log- so i waited about 5 minutes (to 0.2%) had no more errors and cancelled parity parity check and restarted it.  so far no errors on retry (at 0.5%).  I have had intermittent issues in the past where the box will not see any of the drives in the 16 drive and have to be powered down.  My best guess on that is that it was cabling or the connection to a hard drive not being secure.  But that is only a guess.

 

 

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=888

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=896

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=896

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=904

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=904

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=912

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=912

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=920

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=920

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=928

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=928

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=936

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=936

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=944

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=944

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=952

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=952

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=960

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=960

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=968

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=968

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=976

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=976

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=984

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=984

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=992

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=992

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=1000

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=1000

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=1008

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=1008

Oct 25 13:33:24 Tower kernel: md: disk3 read error, sector=1016

Oct 25 13:33:24 Tower kernel: md: disk18 read error, sector=1016

 

after i restarted the parity check 10 minutes in as noted in post above, i got no errors on this parity check.

So, i believe that these read errors are not real, but unraid has listed them 128 on drive 3 and 18 each.  So the array is considered bad.  Will a reboot reset those read errors?  If not how does one reset read errors that they believe to be false?

Will a reboot reset those read errors?

 

Yes