Warning Notification Question

[wgstarks]

I get this notification every time I reboot my server. Can someone tell me what it means?

Event: unRAID Disk 1 SMART health [187]
Subject: Warning [bRUNNHILDE] - reported uncorrect is 22
Description: ST3000DM001-9YN166_Z1F0KHH2 (sdh)
Importance: warning

[dgaschk]

Tools->Diagnostics. Post file.

[wgstarks]

Attached

brunnhilde-diagnostics-20150805-2304.zip

[dgaschk]

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
187 Reported_Uncorrect      0x0032   078   078   000    Old_age   Always       -       22

 

The current VALUE is 78. As long as the value remains above the TRESHold of 0 the drive is OK.

[SSD]

HERE is a good source of information about S.M.A.R.T. attributes.

 

Attribute 187 [Reported_uncorrectable] says that this is a situation where ECC could not correct data read from the disk. My understanding is that drive media at the tiny tolerances involved is written with redundancy such that if can be corrected if the read data is slightly corrupted. (It is considered normal that there would be such corruption and ECC is the means to correct it).

 

Seems by the description that this attribute is triggered if a read cannot be corrected with the ECC information. I am not sure what happens in such a case. I would think that the drive would report a read error back to the OS. unRAID, upon receiving such a read error, would reconstruct the sector and re-write the data to the errant block, and hopefully all would be well.

 

Although I might not jump out of my skin with 22 of these, it certainly doesn't sound too good to me. The drive has degraded the normalized attribute setting sharply (from 100 to 78), apparently dropping it one point per reported_uncorrect, such that once you get to 100, it would mark the drive as failed.

 

I'd run a few parity checks and see if the value is continuing to increase. If it is, suggest you replace the drive.

[wgstarks]

Thanks guys. Guess it's time to start shopping for a new hot spare.

[CHBMB]

Thanks guys. Guess it's time to start shopping for a new hot spare.

 

I have two 4TB Seagate drives that always report this error, but haven't changed for months so I now just ignore them..

[wgstarks]

Thanks guys. Guess it's time to start shopping for a new hot spare.

 

I have two 4TB Seagate drives that always report this error, but haven't changed for months so I now just ignore them..

Thanks. I've been seeing it since upgrading to v6. I'm going to run the parity checks as suggested and keep an eye out for good deals on HD's. Been thinking about upgrading parity to 6TB anyway. Might be a good time.

[itimpi]

Thanks guys. Guess it's time to start shopping for a new hot spare.

 

I have two 4TB Seagate drives that always report this error, but haven't changed for months so I now just ignore them..

Thanks. I've been seeing it since upgrading to v6. I'm going to run the parity checks as suggested and keep an eye out for good deals on HD's. Been thinking about upgrading parity to 6TB anyway. Might be a good time.

Just been trying out the Seagate 8TB Archive drives on my system using them as unRAID parity drives (previously I had 6TB WD Red in that role).  They seem to work as well as the 6TB drives did with similar (if not better) performance..  Thought it was worth mentioning as currently those 8TB drives are cheaper than 6TB drives.