User Tools

This is an old revision of the document!


Mdadm Helpers

How do I replace a disk marked as removed from a linux md raid-5 array

If drives are "hot swap", don't need to power off server/pc...

First identify removed drive:

lsblk
Find serial number, if necessary:
smartctl -a /dev/sdb | grep Serial

If drive is just remved:

mdadm --zero-superblock /dev/sdXn
mdadm /dev/md0 --add /dev/sdXn

If replacing damaged drive, then only:

mdadm /dev/md0 --add /dev/sdXn

Check every drive manually

mdadm --examine /dev/sdX

“not enough to start the array” – error while staring mdadm RAID array

Stop failed array:

mdadm -S /dev/md0

Assemble the pre-existing array:

mdadm -A -f /dev/md0

Test Email Settings

mdadm --monitor --scan --test

If got error: “mdadm: Only one autorebuild process allowed in scan mode, aborting”

mdadm --monitor /dev/md0 --test

Simple Script to read Hdd serial numbers

#!/bin/bash
smartctl -a /dev/sda | grep -i 'Serial';
echo "";
echo "";

for i in {a..z} ; do
echo $i
smartctl -a /dev/sd$i | grep -i 'Serial';
done

A SparesMissing event had been detected on md device

Removing the spares=1 option or just recreating the mdadm.conf from scratch fixes the problem:

/usr/share/mdadm/mkconf > /etc/mdadm/mdadm.conf.copy

Double check new configuration ad then save as mdadm.conf