Can i auto empty and remove a drive on SMART failure?

[TheBuz]

I am transferring all of my data from Stablebit Drivepool.

 

It had a feature where if a drive fails a smart check, it empties the drive and removes it from the pool. (and sends a email)

 

Is there a way i could achieve the same in Unraid?

[testdasi]

I think you can set up email for errors but not emptying out a drive that fails a smart check.

 

I don't think that feature is something desirable to begin with.

• SMART errors can vary from "meh" to "holy crap".
• Even the same error (e.g. reallocated sector) can vary from "meh" (e.g. 1 reallocated sector) to "holy crap" (e.g. increasing number of reallocated sectors within a short time).
• A failed SMART check may also vary from "meh" (e.g. bad controller, fixable without touching the drives) to "holy crap".

So when a disk fails SMART, human assessment is almost always required. Doing something as drastic as emptying out a drive automatically can vary from unnecessary to potentially catastrophic.

 

In addition, Stablebit Drivepool does not support parity natively so I can understand the reasoning behind emptying a failed drive out - essentially trying to rescue as much data as possible. For Unraid with parity though, that step is completely not necessary since data can still be emulated even in the case of a catastrophically failed drive and all you need to do is replace the failed drive (i.e. no need to empty out and then move things back).

 

 

 

 

 

 

 

 

 

[JonathanM]

Not without programming it yourself.

 

I wouldn't advise it though. SMART is not by itself a good indicator of drive replacement, there are too many factors involved, and it's just not smart enough.

 

I would definitely recommend setting up the notification system that unraid has, it will alert you to many anomalies that warrant further investigation. Normally what you do with a real drive failure in unraid is to replace the drive with an equal or larger drive, up to the size of your smallest parity drive. It's not recommended to try to move data around while the array is in a degraded state.

[Frank1940]

1 minute ago, jonathanm said:

Normally what you do with a real drive failure in unraid is to replace the drive with an equal or larger drive, up to the size of your smallest parity drive. It's not recommended to try to move data around while the array is in a degraded state.

+1  This should be your first approach to addressing any drive failure.  The vast majority of the time, there will be no loss of data.  (My personal approach at the first sign of a drive related issue is to replace the drive and rebuilt the data on the replacement drive.  Then I troubleshoot the drive to see exactly what the issue is...)

 

3 minutes ago, jonathanm said:

 

I would definitely recommend setting up the notification system that unraid has, it will alert you to many anomalies that warrant further investigation.

+1  Using the notification system will alert you to problems early enough before they escalate into catastrophic events. BUT, you should also have a secure off-site storage of data that is irreplaceable!  Fires, Floods, Earthquakes, Hurricanes, Tornadoes, Lightning, Robbery, Malware are real issues that have to be considered in any data recovery and protection plan.