Advice on possible faulty disk

[skitzor]

Hi all,

Recently Unraid notified me there were some read errors on one of my disks. I have attached the smart log for the disk.

Shortly after I noticed them, the scheduled parity check began, it was showing very slow speed (<1MB/s) and read errors were increasing. So I stopped it and rebooted. Since then there have been more read errors (although the disk would rarely be used, so not much chance for it to to show significant errors). Unraid dashboard reports 10k reads to disk, with only 72 errors.

 

Version: 6.8.2 2020-01-26

Disk: MG04SCA60EE (SAS)

SAS controller: Broadcom / LSI SAS3008 PCI-Express Fusion-MPT SAS-3 (rev 02)

 

One concern is that the SMART report actually shows 2 write errors. I have seen people say that if it's just read errors then it could just be a cabling issue.

Other disks appear to be fine so presumably not a controller issue.

 

Normally I would just proceed with re-seating cable, running ext SMART, and replacing disk if that didn't work etcetc , but unfortunately I am 800km away from the server.

I have a friend that can open it up and be my hands+feet, but they have little experience with computer internals, so I wanted to try and reduce the amount of mucking around I ask them to do.

 

I have seen threads that recommend clearing the disk and doing something on it a few times, and others that say run extended SMART. This is my first issue with an unraid array, and I didn't want to jump the gun, so thought I would post and get some specific advice for this situation so I can get a plan together that would reduce the amount of back and forward for my friend.

 

Thanks!

syslog_20201130.txt MG04SCA60EE_26W0A0D1FWWB_350000396d8325675-20201130-1106.txt

[JorgeB]

SMART is failing and the disk should be replaced.

[skitzor]

ok thanks, will go ahead with that.