Wanted to validate if Unraid uses a standard Raid 6 type mechanism (typically Reed-Solomon coding) under the hood when using dual parity? 

 

The question is specifically to address if Dual parity would correct a bit flip or corruption in the resilvering (parity check) process. If not then would dual parity only help in the case of multiple drive failures?

 

Hoping to hear from someone from Unraid with an official answer.

 

Thank you

5 hours ago, mindedc said:

Wanted to validate if Unraid uses a standard Raid 6 type mechanism (typically Reed-Solomon coding) under the hood when using dual parity? 

It uses this:

https://mirrors.edge.kernel.org/pub/linux/kernel/people/hpa/raid6.pdf

 

5 hours ago, mindedc said:

The question is specifically to address if Dual parity would correct a bit flip or corruption in the resilvering (parity check) process. If not then would dual parity only help in the case of multiple drive failures?

The latter, IIRC in part because of this from the link above:

 

Quote

Finally, as a word of caution it should be noted that RAID-6 by itself cannot (in the general case) even detect, never mind recover from, dual-disk corruption. If two disks are corrupt in the same byte positions, the above algorithm will (again, in the general case) introduce additional data corruption by corrupting a third drive.

 

I had meant a corruption of a single drive which the whitepaper clearly indicates is detected and repaired in a parity recalculation. After the last sentence you quoted they go on to state that such detection is possible but the code doesn't support more than two drives right now which is completely reasonable. 

 

Thank you so much for your support.