While configuring RAID it is always advised to add a spare partition to your raid device so that in case of any hard disk failure the spare partition can be utilized
Lets add a new virtual hard disk to our machine.
To keep this article strictly on the roadmap I will assume that you already have created a new partition and changed the partition type from default to "Linux RAID". If not you can follow my last article where I had explained in detail to do the same.
So we will now just add the new partition to our raid device
# mdadm --manage /dev/md0 --add /dev/sdd1
mdadm: added /dev/sdd1Once added let us look at the details of your raid device
# mdadm --detail /dev/md0
/dev/md0:
Version : 1.2
Creation Time : Fri Sep 19 23:02:52 2014
Raid Level : raid1
Array Size : 5233024 (4.99 GiB 5.36 GB)
Used Dev Size : 5233024 (4.99 GiB 5.36 GB)
Raid Devices : 2
Total Devices : 3
Persistence : Superblock is persistent
Update Time : Sat Sep 20 02:06:39 2014
State : clean
Active Devices : 2
Working Devices : 3
Failed Devices : 0
Spare Devices : 1
Name : test2.example:0 (local to host test2.example)
UUID : 5a463788:9bf2659a:09d1c73a:9adcbbbd
Events : 76
Number Major Minor RaidDevice State
2 8 17 0 active sync /dev/sdb1
1 8 33 1 active sync /dev/sdc1
3 8 49 - spare /dev/sdd1So now as it says there are total 3 devices out of which 2 devices are used for RAID whereas one is kept as spare device.
Well our work is not yet done as this article is all about growing the size of the raid array so lets do that
# mdadm --grow --raid-devices=3 /dev/md0
raid_disks for /dev/md0 set to 3Once done run the below command
# cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdd1[3] sdb1[2] sdc1[1]
5233024 blocks super 1.2 [3/2] [UU_]
[======>..............] recovery = 34.3% (1800064/5233024) finish=0.2min speed=225008K/sec unused devices: <none> As you can see a recovery process has started and 3rd device can be seen under the array. Well this is not exactly a recovery process instead it is performing an action of adding the device to the group and performing mirroring.
You can see the same in details using the below command
# mdadm --detail /dev/md0
/dev/md0:
Version : 1.2
Creation Time : Fri Sep 19 23:02:52 2014
Raid Level : raid1
Array Size : 5233024 (4.99 GiB 5.36 GB)
Used Dev Size : 5233024 (4.99 GiB 5.36 GB)
Raid Devices : 3
Total Devices : 3
Persistence : Superblock is persistent
Update Time : Sat Sep 20 02:09:22 2014
State : clean, degraded, recovering
Active Devices : 2
Working Devices : 3
Failed Devices : 0
Spare Devices : 1
Rebuild Status : 53% complete
Name : test2.example:0 (local to host test2.example)
UUID : 5a463788:9bf2659a:09d1c73a:9adcbbbd
Events : 88
Number Major Minor RaidDevice State
2 8 17 0 active sync /dev/sdb1
1 8 33 1 active sync /dev/sdc1
3 8 49 2 spare rebuilding /dev/sdd1Once the recovery process is completed you should be able to see 3 active devices under the raid array
# mdadm --detail /dev/md0
/dev/md0:
Version : 1.2
Creation Time : Fri Sep 19 23:02:52 2014
Raid Level : raid1
Array Size : 5233024 (4.99 GiB 5.36 GB)
Used Dev Size : 5233024 (4.99 GiB 5.36 GB)
Raid Devices : 3
Total Devices : 3
Persistence : Superblock is persistent
Update Time : Sat Sep 20 02:09:35 2014
State : clean
Active Devices : 3
Working Devices : 3
Failed Devices : 0
Spare Devices : 0
Name : test2.example:0 (local to host test2.example)
UUID : 5a463788:9bf2659a:09d1c73a:9adcbbbd
Events : 99
Number Major Minor RaidDevice State
2 8 17 0 active sync /dev/sdb1
1 8 33 1 active sync /dev/sdc1
3 8 49 2 active sync /dev/sdd1
What would happen if any of the hard disk crashes?
Let us try it out and see the output by ourself
Now I will manually fail /dev/sdb1
# mdadm --fail /dev/md0 /dev/sdb1
mdadm: set /dev/sdb1 faulty in /dev/md0Check the detailed output once again
# mdadm --detail /dev/md0
/dev/md0:
Version : 1.2
Creation Time : Fri Sep 19 23:02:52 2014
Raid Level : raid1
Array Size : 5233024 (4.99 GiB 5.36 GB)
Used Dev Size : 5233024 (4.99 GiB 5.36 GB)
Raid Devices : 3
Total Devices : 3
Persistence : Superblock is persistent
Update Time : Sat Sep 20 02:13:31 2014
State : clean, degraded
Active Devices : 2
Working Devices : 2
Failed Devices : 1
Spare Devices : 0
Name : test2.example:0 (local to host test2.example)
UUID : 5a463788:9bf2659a:09d1c73a:9adcbbbd
Events : 101
Number Major Minor RaidDevice State
0 0 0 0 removed
1 8 33 1 active sync /dev/sdc1
3 8 49 2 active sync /dev/sdd1
2 8 17 - faulty /dev/sdb1Now since we have one faulty device, we can just remove it as it is of no use
# mdadm --remove /dev/md0 /dev/sdb1
mdadm: hot removed /dev/sdb1 from /dev/md0Resize the raid array and update the raid devices to the available no. of active partitions i.e. 2
# mdadm --grow --raid-devices=2 /dev/md0
raid_disks for /dev/md0 set to 2Verify the raid device output
# mdadm --detail /dev/md0
/dev/md0:
Version : 1.2
Creation Time : Fri Sep 19 23:02:52 2014
Raid Level : raid1
Array Size : 5233024 (4.99 GiB 5.36 GB)
Used Dev Size : 5233024 (4.99 GiB 5.36 GB)
Raid Devices : 2
Total Devices : 2
Persistence : Superblock is persistent
Update Time : Sat Sep 20 02:14:39 2014
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Name : test2.example:0 (local to host test2.example)
UUID : 5a463788:9bf2659a:09d1c73a:9adcbbbd
Events : 109
Number Major Minor RaidDevice State
1 8 33 0 active sync /dev/sdc1
3 8 49 1 active sync /dev/sdd1So we are back with 2 active raid devices without loosing any single data.
I hope the article was helpful.
Related Articles
How to detect new hard disk attached without rebooting in Linux
How to detect new NIC/Ethernet card without rebooting in Linux
Taking Backup of Hard Disk
Disk Attachment Technology FC vs SAS vs iSCSI