In article <[email protected]>, Louis Guillaume <[email protected]> wrote: >I recently had a component fail in a raidframe RAID-1 array... > >wd1a: error reading fsbn 267569088 of 267569088-267569119 (wd1 bn >267571136; cn 265447 tn 8 sn 56) >wd1: (obsolete (address mark not found)) >raid1: IO Error. Marking /dev/wd1a as failed. > >So I did some inspection on the disk with atactl and smartctl and it >seems like maybe there were some bad blocks but overall the disk was ok. >Fine. So I decided to reconstruct with this same drive. > >It's been about 3 days and it's only at 7%. It's painfully slow. Bad >blocks are found, but they're being remapped as expected and there are >only a handful so far. > >At this rate though, a replacement disk will be here before the >reconstruction is done. So I'd like to cancel the reconstruction. > >Is there a safe way to do that?
I think shutting down and rebooting will do it. christos
