The timeline is as follows:
- I've got 5 2TB hdd's in raid5 array
- Added another 2TB hdd, using mdadm --add /dev/md125 /dev/sdf1
- Started converting array using mdadm --grow /dev/md125 --level=6 --raid-devices=6 --backup-file=/mnt/rb/md125.backup
- Then /dev/hdc1 fails at about 0.8% of rebuild, so it's [UUU_U_] now.
- Decided to add hot-spare to the array: mdadm --add /dev/md125 /dev/sdg1
- Now md125_raid5 eats 100% cpu and all disk activity on raid drives is stalled
Any ideas on what to do?