Hey all. 

 

I upgraded my 3tb parity disk with a 10tb disk. Then used the old parity drive to replace a smaller, older disk5. 

I'm now getting SMART errors on disk5 (the old parity drive). 

I ran the extended test, but I'm not familiar with the errors enough to know if I should be concerned or not. Lots of "power-on lifetime" errors and it's a WD drive, I heard they get a lot of errors that other drives do not. 

 

My biggest concern is being able to rebuild if any drive fails. 

Tonight I'll probably scatter all the data off disk5 until I know what the disk needs. 

 

Any help is appreciated. 

extended-smart-test-disk5.txt doctorhands-diagnostics-20200427-1703.zip

Edited April 28, 20206 yr by 1trkmind

Extended SMART test failed = failed disk

 

SMART Extended Self-test Log Version: 1 (1 sectors)
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Extended offline    Completed: read failure       10%     28170         5845199728

 

 

3 hours ago, johnnie.black said:

Extended SMART test failed = failed disk

 

SMART Extended Self-test Log Version: 1 (1 sectors)
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Extended offline    Completed: read failure       10%     28170         5845199728

 

 

The overall result of the test is a "Pass". (SMART overall-health self-assessment test result: PASSED)

Are you saying that if any individual test component fails, the entire disk is bad? 

 

2 minutes ago, 1trkmind said:

The overall result of the test is a "Pass". (SMART overall-health self-assessment test result: PASSED)

That does not take into consideration what SMART tests have been run - just the attribute values.

 

2 minutes ago, 1trkmind said:

Are you saying that if any individual test component fails, the entire disk is bad? 

Typically yes  

Thanks guys. Unraid is so picky compared to windows on disk health 😉

11 minutes ago, 1trkmind said:

The overall result of the test is a "Pass". (SMART overall-health self-assessment test result: PASSED)

That's mostly worthless as a way to access a disk health, it only changes if there's a FAILING NOW SMART attribute.

I guess it's time to shuck another 10tb WD Easystore. 

On 4/28/2020 at 7:53 AM, 1trkmind said:

Thanks guys. Unraid is so picky compared to windows on disk health 😉

Windows only alerts you to an issue when there is data attempted to be read from a bad spot. Even then there typically isn't an error shown, you have to dig through the logs to find it. Unraid uses the entire disk capacity for parity protection, so any disk read error can cause issues rebuilding another failed disk.

 

Windows will happily continue to use a disk until it dies completely, sometimes with no indication you are in trouble until you get a failure to boot.

 

I have worked on countless machines where the users would have much rather gotten some warning than be told their only option was to send the disk away to a clean room recovery center.

 

Windows leaves you blissfully ignorant of a bad situation.