Mdadm: Difference between revisions
No edit summary |
|||
Line 38: | Line 38: | ||
cat /sys/block/md0/md/mismatch_cnt | cat /sys/block/md0/md/mismatch_cnt | ||
</pre> | </pre> | ||
==Changing Raid Levels== | |||
Note: I haven't tested this. | |||
===Raid 1 to Raid 5=== | |||
https://unix.stackexchange.com/questions/610100/is-it-possible-to-keep-data-in-procedure-raid1-growing-to-raid5 | |||
===Raid 5 to Raid 6=== |
Revision as of 22:32, 30 November 2024
Mdadm is used to create linux MD raids.
Usage
# Check the status of /dev/md0
mdadm -D /dev/md0
Create a raid array
See https://www.tecmint.com/create-raid-6-in-linux/
First use gdisk to create a linux raid partition (FD00) on all your disks. Note that is purely convention and is optional.
Then run the following to create /dev/md0.
mdadm --create /dev/md0 --level=6 --raid-devices=4 /dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdd1
# Now you have a block device /dev/md0 on which you can add LUKS, LVM, or a filesystem.
Add hot spare
See https://tuxfixer.com/how-to-add-hot-spare-volume-to-the-existing-mdadm-software-raid-array/
mdadm --add /dev/md0 $DEVICE
Scrubs
On Ubuntu, md will automatically scrub on the first Sunday of every month using a systemd timer.
During a scrub, you can check the status with
cat /proc/mdstat
You can check for bad blocks using
cat /sys/block/md0/md/mismatch_cnt
Changing Raid Levels
Note: I haven't tested this.