Disk Fail -- wondering if I need to replace or just re-enable

[andrew207]

Disk 2 has been auto-disabled for about a week. Seems like I just got a bad disk, it's only about 9 months old 8TB IronWolf.

 

Diag attached, also SMART report for disk that has died (Disk 2). The SMART numbers look very severe, but I don't fully understand their implication. In particular:

 

ID# ATTRIBUTE_NAME          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate     POSR--   082   064   044    -    164804696

 

I'd like to know whether the disk is broken enough that I should immediately send it off for warranty, or if I can just enable it again and it'll work fine for another few months.

 

Thanks all

 

 

 

rack-smart-20190407-1138.zip

rack-diagnostics-20190407-1142.zip

[Squid]

27 minutes ago, andrew207 said:

Diag attached, also SMART report for disk that has died (Disk 2). The SMART numbers look very severe, but I don't fully understand their implication. In particular:

Nope.  Raw Error Rate on Seagate drives is a manufacturer specific attribute, and is basically meaningless to humans. 

 

Doesn't look like when the drive was disabled is present within the syslog, so can't really tell why it was (beyond that a write failed to it), but since SMART is fine then odds on it is the usual problem of a crappy SATA connection, or improperly tidying up cabling etc.

 

40 minutes ago, andrew207 said:

Disk 2 has been auto-disabled for about a week

You do realize that if something else happened during that week then you would have lost data.

 

As an aside,

Quote

Serial Number:    [redacted]

The s/n is also in the filename (unredacted), and in the diagnostics everywhere.  Redacting S/N's is pointless because there's nothing anyone can do with them.  It is kind of like spray-painting your windshield over the VIN number.  All other info which does mean something (share names, filenames on the server, etc) are automatically anonymized.