1

I am aware of the problem when using standard consumer-grade drives with RAID levels 5 and 6; if a drive takes too long trying to read and/or repair a bad sector, it will be dropped by a hardware RAID controller because the controller thinks the drive is dead.

ERC/TLER/CCTL is supposed to prevent this by halting the failed I/O attempt, informing the hardware RAID controller of the problem and allowing the bad sector to be rebuilt from parity data (in RAID 5 and RAID 6). In RAID 1, the exact same data is shared across all drives, and thus does not use parity data. I am not sure how this changes the rebuild process of the array.

Specifically, does using ERC/TLER/CCTL-capable drives provide any benefit for a RAID 1 configuration?

Improve this question

1 Answer 1

Reset to default
1

It provides the same advantage for RAID1 as for RAID5/6: A read from a sector with an error will be aborted in shorter time. Thus attempted read for that sector will not cause a long delay. This can be quite nice if that RAID1 holds your OS and the whole OS freezes as a result.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .