企业经常用到的RAID+LVM管理模式：
用到的软件：VMware Workstation 10
系统环境：
|
1
2
3
4
5
|
[root@localhost~]
# cat /etc/centos-release
CentOSLinux release 7.0.1406 (Core)
[root@localhost~]
# rpm -qi mdadm
Name : mdadm
Version : 3.2.6
|
RAID 1+0方案
简介
先对磁盘做mirror，然后对整个mirror组做条带化；比如8块盘需要分成4个基组，每个基组2块盘；每个基组先做raid1，再做raid0，4条条带化。
所以：
允许所有磁盘基组中的磁盘各损坏一个，但是不允许同一基组中的磁盘同时有坏的。
优点：数据安全性好，只要不是1个条带上的2个硬盘同时坏，没有问题，还可以继续跑数据。数据恢复快。
缺点 ：写性能稍微比RAID 0+1 差（读性能一样）
我的思路
在新添加的硬盘上划分4个1G的区域分别为1,2,3,5
每两个转化为一个RAID 1 卷 /dev/md{1,2}
将这两个RAID 1 卷转化为一个RAID 0卷/dev/md0
将RAID 0转化为pv，vg
之后通过lv创建自己的空间
这个空间就具有RAID 1+0的性质
详细实施过程
我们这里添加一块用作实验的硬盘2
我们不想重启来读取新添加的硬盘，我们输入下面命令进行扫描
|
1
2
3
4
5
6
7
8
|
[root@localhost~]
# echo "- - -" > /sys/class/scsi_host/host0/scan
[root@localhost~]
# echo "- - -" > /sys/class/scsi_host/host1/scan
[root@localhost~]
# echo "- - -" > /sys/class/scsi_host/host2/scan
[root@localhost~]
# fdisk -l
Disk
/dev/sdb
: 21.5 GB, 21474836480 bytes, 41943040sectors
Units= sectors of 1 * 512 = 512 bytes
Sectorsize (logical
/physical
): 512 bytes / 512 bytes
I
/Osize
(minimum
/optimal
): 512 bytes / 512 bytes
|
发现新添加的sdb
下面开始具体的实验过程
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
|
#fdisk/dev/sdb
Command(m
for
help): n
Partitiontype:
p primary (0 primary, 0 extended, 4
free
)
e extended
Select(default p):
Usingdefault response p
Partitionnumber (1-4, default 1):
Firstsector (2048-41943039, default 2048):
Usingdefault value 2048
Lastsector, +sectors or +size{K,M,G} (2048-41943039, default 41943039): +1G
Partition1 of
type
Linux and of size 1 GiB is
set
Command(m
for
help): n
Partitiontype:
p primary (1 primary, 0 extended, 3
free
)
e extended
Select(default p):
Usingdefault response p
Partitionnumber (2-4, default 2):
Firstsector (2099200-41943039, default 2099200):
Usingdefault value 2099200
Lastsector, +sectors or +size{K,M,G} (2099200-41943039, default 41943039): +1G
Partition2 of
type
Linux and of size 1 GiB is
set
Command(m
for
help): n
Partitiontype:
p primary (2 primary, 0 extended, 2
free
)
e extended
Select(default p):
Usingdefault response p
Partitionnumber (3,4, default 3):
Firstsector (4196352-41943039, default 4196352):
Usingdefault value 4196352
Lastsector, +sectors or +size{K,M,G} (4196352-41943039, default 41943039): +1G
Partition3 of
type
Linux and of size 1 GiB is
set
Command(m
for
help): n
Partitiontype:
p primary (3 primary, 0 extended, 1
free
)
e extended
Select (default e):
Usingdefault response e
Selectedpartition 4
Firstsector (6293504-41943039, default 6293504):
Usingdefault value 6293504
Lastsector, +sectors or +size{K,M,G} (6293504-41943039, default 41943039):
Usingdefault value 41943039
Partition4 of
type
Extended and of size 17 GiB is
set
|
将剩余的所有空间划给扩展区，之后再划分区间用l命令即逻辑盘区
此后的区间号就从5开始往后加
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
|
Command(m
for
help): n
Allprimary partitions are
in
use
Addinglogical partition 5
Firstsector (6295552-41943039, default 6295552):
Usingdefault value 6295552
Lastsector, +sectors or +size{K,M,G} (6295552-41943039, default 41943039): +1G
Partition5 of
type
Linux and of size 1 GiB is
set
Command(m
for
help): p
Disk
/dev/sdb
: 21.5 GB, 21474836480 bytes, 41943040 sectors
Units= sectors of 1 * 512 = 512 bytes
Sectorsize (logical
/physical
): 512 bytes / 512 bytes
I
/Osize
(minimum
/optimal
): 512 bytes / 512 bytes
Disklabel
type
: dos
Diskidentifier: 0x8e9a9326
Device Boot Start End Blocks Id System
/dev/sdb1
2048 2099199 1048576 83 Linux
/dev/sdb2
2099200 4196351 1048576 83 Linux
/dev/sdb3
4196352 6293503 1048576 83 Linux
/dev/sdb4
6293504 41943039 17824768 5 Extended
/dev/sdb5
6295552 8392703 1048576 83 Linux
/dev/sdb6
8394752 10491903 1048576 83 Linux
[root@localhost ~]
# mdadm -C /dev/md1-a yes -l 1 -n 2 /dev/sdb{1,2}
mdadm:Note: this array has metadata at the start and
may not be suitable as a boot device. If you plan to
store
'/boot'
on this device please ensurethat
your boot-loader understands md
/v1
.xmetadata, or use
--metadata=0.90
Continuecreating array?
Continuecreating array? (y
/n
) y
mdadm:Defaulting to version 1.2 metadata
mdadm:array
/dev/md1
started.
[root@localhost ~]
# mdadm -C /dev/md2 -a yes-l 1 -n 2 /dev/sdb{3,5}
mdadm:Note: this array has metadata at the start and
may not be suitable as a boot device. If you plan to
store
'/boot'
on this device please ensurethat
your boot-loader understands md
/v1
.xmetadata, or use
--metadata=0.90
Continuecreating array?
Continuecreating array? (y
/n
) y
mdadm:Defaulting to version 1.2 metadata
mdadm:array
/dev/md2
started.
[root@localhost ~]
# mdadm -C /dev/md0 -a yes-l 1 -n 2 /dev/md{1,2}（将RAID1转化为RAID0）
mdadm:Note: this array has metadata at the start and
may not be suitable as a boot device. If you plan to
store
'/boot'
on this device please ensurethat
your boot-loader understands md
/v1
.xmetadata, or use
--metadata=0.90
Continuecreating array?
Continuecreating array? (y
/n
) y
mdadm:Defaulting to version 1.2 metadata
mdadm:array
/dev/md0
started.
[root@localhost ~]
# pvcreate /dev/md0（变PV）
Physical volume
"/dev/md0"
successfully created
[root@localhost ~]
# vgcreate first/dev/md0（变VG）
Volume group
"first"
successfullycreated
[root@localhost~]
# lvcreate -L 1G -n data1 first
Volume group
"first"
hasinsufficient
free
space (255 extents): 256 required.
[root@localhost ~]
# lvcreate -L +500M-n data1 first（划分自己想要的区域大小）
Logical volume
"data1"
created
[root@localhost ~]
# mkfs.ext4/dev/first/data1（格式化）
mke2fs1.42.9 (28-Dec-2013)
Filesystemlabel=
OStype: Linux
Blocksize=1024 (log=0)
Fragmentsize=1024 (log=0)
Stride=0blocks, Stripe width=0 blocks
128016inodes, 512000 blocks
25600blocks (5.00%) reserved
for
the super user
Firstdata block=1
Maximumfilesystem blocks=34078720
63block
groups
8192blocks per group, 8192 fragments per group
2032inodes per group
Superblockbackups stored on blocks:
8193, 24577, 40961, 57345, 73729,204801, 221185, 401409
Allocatinggroup tables:
done
Writinginode tables:
done
Creatingjournal (8192 blocks):
done
Writingsuperblocks and filesystem accounting information:
done
[root@localhost~]
# mkdir /lvm1
[root@localhost~]
# mount /dev/first/data1 /lvm1
[root@localhost~]
# cd /lvm1
[root@localhostlvm1]
# ll
total12
drwx------2 root root 12288 Aug 12 21:21 lost+found
|
挂载后即可对划分出的新空间进行使用。
举例
这里举个例子，20个硬盘
做RAID 1+0，共10个条带，每个条带2个硬盘做MIRROR，如果坏了1个硬盘，没关系，其它19个硬盘还要同时工作，只要不是坏在一个MIRROR里面的，就没事。
建议，硬盘很多时，同时坏的几率就比较大，建议使用安全系数高的RAID 1+0,宁愿损失点性能（其实差不多)。
后语：
新手，如有错误，欢迎指正。
本文转自 紫色的茶碗 51CTO博客，原文链接：http://blog.51cto.com/chawan/1540190，如需转载请自行联系原作者
