Re: En théorie ....

Voici le test :

On commence par créer les images disque et initialiser les peripheriques loop:

for i in 1 2 3 4 5 ; do dd if=/dev/zero of=test${i}.img bs=1M count=1024; sudo losetup /dev/loop${i} ./test${i}.img ; done
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 0.828014 s, 1.3 GB/s
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 2.93688 s, 366 MB/s
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 3.44726 s, 311 MB/s
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 3.44369 s, 312 MB/s
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 5.11845 s, 210 MB/s

Ensuite on crée le device

$ sudo mdadm --create /dev/md0 -n 3 -l 5 /dev/loop1 /dev/loop2 /dev/loop3
mdadm: Defaulting to version 1.2 metadata
mdadm: array /dev/md0 started.

On crée le filesystem:

$ sudo mkfs -T ext4 /dev/md0
mke2fs 1.42.9 (28-Dec-2013)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=128 blocks, Stripe width=256 blocks
131072 inodes, 523776 blocks
26188 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=536870912
16 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
 32768, 98304, 163840, 229376, 294912
Allocating group tables: done
Writing inode tables: done
Creating journal (8192 blocks): done
Writing superblocks and filesystem accounting information: done

On monte le filesystem:

$ sudo mount /dev/md0 /mnt && df /mnt
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/md0 2029392 6144 1902112 1% /mnt

On ajoute les devices :

$ sudo mdadm --manage /dev/md0 --add /dev/loop5
mdadm: added /dev/loop5
$ sudo mdadm --manage /dev/md0 --add /dev/loop4
mdadm: added /dev/loop4

extension de l'array sur les deux disques :

$ sudo mdadm --grow /dev/md0 --raid-devices=5
mdadm: Need to backup 2048K of critical section..

On vérifie l'état de la reconstruction :

$ cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
md0 : active raid5 loop4[5] loop5[4] loop3[3] loop2[1] loop1[0]
 2095104 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
 [======>..............] reshape = 34.1% (358400/1047552) finish=0.3min speed=29866K/sec
unused devices: <none>
$ cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
md0 : active raid5 loop4[5] loop5[4] loop3[3] loop2[1] loop1[0]
 2095104 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
 [=============>.......] reshape = 69.3% (727040/1047552) finish=0.1min speed=33047K/sec
unused devices: <none>
$ cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
md0 : active raid5 loop4[5] loop5[4] loop3[3] loop2[1] loop1[0]
 2095104 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
 [===================>.] reshape = 95.2% (998400/1047552) finish=0.0min speed=34427K/sec
unused devices: <none>
$ cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
md0 : active raid5 loop4[5] loop5[4] loop3[3] loop2[1] loop1[0]
 2095104 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
 [===================>.] reshape = 95.2% (998400/1047552) finish=0.0min speed=28525K/sec
unused devices: <none>
$ cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
md0 : active raid5 loop4[5] loop5[4] loop3[3] loop2[1] loop1[0]
 2095104 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
 [===================>.] reshape = 95.3% (998912/1047552) finish=0.0min speed=25613K/sec
unused devices: <none>
$ cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
md0 : active raid5 loop4[5] loop5[4] loop3[3] loop2[1] loop1[0]
 4190208 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
unused devices: <none>

Une fois que c'est terminé, on redimensionne le fs (j'ai fait un df avant et après le resize2fs pous s'assurer que le fs est bien étendu):

$ df /mnt
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/md0 2029392 6144 1902112 1% /mnt
$ resize2fs /dev/md0
resize2fs 1.42.9 (28-Dec-2013)
open: Permission denied while opening /dev/md0
$ df /mnt
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/md0 2029392 6144 1902112 1% /mnt
$ sudo resize2fs /dev/md0
resize2fs 1.42.9 (28-Dec-2013)
Filesystem at /dev/md0 is mounted on /mnt; on-line resizing required
old_desc_blocks = 1, new_desc_blocks = 1
The filesystem on /dev/md0 is now 1047552 blocks long.
$ df /mnt
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/md0 4091584 8184 3878544 1% /mnt