软件磁盘阵列(Software RAID)

来源：互联网发布：sql server denty_rank 编辑：程序博客网时间：2024/06/08 19:10

一块新增的7G硬盘，将其分成6个分区，每个分区1G空间。sdb{1,2,3,5,6,7}就是我们接下来实现RAID所要用到的分区。用sdb{1,2,3,5}四个分区组成RAID5，利用sdb6作为sparedisk，sdb7作为备用分区。

一、 mdadm命令创建RAID

1.mdadm命令用于创建软件磁盘阵列，命令格式为：mdadm [模式选项] [RAID设备名] [子选项…] [组件设备名…]

--create为mdadm的创建模式

md0为创建的磁盘阵列名

--level=[015]设置磁盘阵列的等级，我们这里设置为5

--raid-devices=N表明使用几块磁盘作为磁盘阵列使用的设备，我们这里使用了4块，也就是sdb{1,2,3,5}

--spare-devices=N表明使用几块磁盘作为备用设备，我们这里使用了一块，也就是sdb6

2.查看md0阵列的详细信息

[root@minimal ~]# mdadm --detail /dev/md0/dev/md0:        Version : 1.2  Creation Time : Wed Oct 21 11:36:44 2015     Raid Level : raid5     Array Size : 3176448 (3.03 GiB 3.25 GB)  Used Dev Size : 1058816 (1034.17 MiB 1084.23 MB)   Raid Devices : 4  Total Devices : 5    Persistence : Superblock is persistent    Update Time : Wed Oct 21 11:36:58 2015          State : clean  Active Devices : 4Working Devices : 5 Failed Devices : 0  Spare Devices : 1         Layout : left-symmetric     Chunk Size : 512K           Name : minimal:0  (local to host minimal)           UUID : f99d657f:0db77911:9870b07b:92c4c1d9         Events : 18    Number   Major   Minor   RaidDevice State       0       8       17        0      active sync   /dev/sdb1       1       8       18        1      active sync   /dev/sdb2       2       8       19        2      active sync   /dev/sdb3       5       8       21        3      active sync   /dev/sdb5       4       8       22        -      spare   /dev/sdb6

3.格式化并且挂载RAID

4.记录阵列信息，防止系统重启后阵列丢失

[root@minimal ~]# mdadm -Ds >> /etc/mdadm.conf

5.查看配置文件里面的信息

[root@minimal ~]# cat /etc/mdadm.confARRAY /dev/md0 metadata=1.2 spares=1 name=minimal:0 UUID=f99d657f:0db77911:9870b07b:92c4c1d9

二、模拟磁盘阵列出错并且恢复

1.设置sdb5出错

[root@minimal ~]# mdadm --manage /dev/md0 --fail /dev/sdb5

2.查看磁盘阵列当前信息

[root@minimal ~]# mdadm --detail /dev/md0/dev/md0:        Version : 1.2  Creation Time : Wed Oct 21 11:36:44 2015     Raid Level : raid5     Array Size : 3176448 (3.03 GiB 3.25 GB)  Used Dev Size : 1058816 (1034.17 MiB 1084.23 MB)   Raid Devices : 4  Total Devices : 5    Persistence : Superblock is persistent    Update Time : Wed Oct 21 12:02:50 2015          State : clean, degraded, recovering  Active Devices : 3Working Devices : 4 Failed Devices : 1  Spare Devices : 1         Layout : left-symmetric     Chunk Size : 512K Rebuild Status : 61% complete           Name : minimal:0  (local to host minimal)           UUID : f99d657f:0db77911:9870b07b:92c4c1d9         Events : 29    Number   Major   Minor   RaidDevice State       0       8       17        0      active sync   /dev/sdb1       1       8       18        1      active sync   /dev/sdb2       2       8       19        2      active sync   /dev/sdb3       4       8       22        3      spare rebuilding   /dev/sdb6       5       8       21        -      faulty   /dev/sdb5

快速输入mdadm命令可以看到sdb6正在重建数据。如果过一段时间再使用mdadm查看的话会产生下面的结果。

[root@minimal ~]# mdadm --detail /dev/md0/dev/md0:        Version : 1.2  Creation Time : Wed Oct 21 11:36:44 2015     Raid Level : raid5     Array Size : 3176448 (3.03 GiB 3.25 GB)  Used Dev Size : 1058816 (1034.17 MiB 1084.23 MB)   Raid Devices : 4  Total Devices : 5    Persistence : Superblock is persistent    Update Time : Wed Oct 21 12:02:56 2015          State : clean  Active Devices : 4Working Devices : 4 Failed Devices : 1  Spare Devices : 0         Layout : left-symmetric     Chunk Size : 512K           Name : minimal:0  (local to host minimal)           UUID : f99d657f:0db77911:9870b07b:92c4c1d9         Events : 37    Number   Major   Minor   RaidDevice State       0       8       17        0      active sync   /dev/sdb1       1       8       18        1      active sync   /dev/sdb2       2       8       19        2      active sync   /dev/sdb3       4       8       22        3      active sync   /dev/sdb6       5       8       21        -      faulty   /dev/sdb5

3.删除出错的磁盘并且加入新的磁盘

[root@minimal ~]# mdadm --manage /dev/md0 --remove /dev/sdb5 --add /dev/sdb7

命令执行成功后状态：

4.查看当前磁盘阵列信息

[root@minimal ~]# mdadm --detail /dev/md0/dev/md0:        Version : 1.2  Creation Time : Wed Oct 21 11:36:44 2015     Raid Level : raid5     Array Size : 3176448 (3.03 GiB 3.25 GB)  Used Dev Size : 1058816 (1034.17 MiB 1084.23 MB)   Raid Devices : 4  Total Devices : 5    Persistence : Superblock is persistent    Update Time : Wed Oct 21 12:08:34 2015          State : clean  Active Devices : 4Working Devices : 5 Failed Devices : 0  Spare Devices : 1         Layout : left-symmetric     Chunk Size : 512K           Name : minimal:0  (local to host minimal)           UUID : f99d657f:0db77911:9870b07b:92c4c1d9         Events : 39    Number   Major   Minor   RaidDevice State       0       8       17        0      active sync   /dev/sdb1       1       8       18        1      active sync   /dev/sdb2       2       8       19        2      active sync   /dev/sdb3       4       8       22        3      active sync   /dev/sdb6       5       8       23        -      spare   /dev/sdb7

三、磁盘阵列扩容

1.向raid5中添加新设备（分区或者磁盘）

[root@minimal ~]# mdadm /dev/md0 --add /dev/sdb5mdadm: added /dev/sdb5

2.改变raid5工作盘的个数

[root@minimal ~]# mdadm --grow /dev/md0 --raid-devices=5mdadm: Need to backup 6144K of critical section..

3.调整文件系统大小

[root@minimal ~]# resize2fs /dev/md0resize2fs 1.41.12 (17-May-2010)Filesystem at /dev/md0 is mounted on /mnt/raid; on-line resizing requiredold desc_blocks = 1, new_desc_blocks = 1Performing an on-line resize of /dev/md0 to 1058816 (4k) blocks.The filesystem on /dev/md0 is now 1058816 blocks long.

使用df查看挂载设备，会发现md0已经增加了1G的空间

[root@minimal ~]# df -hFilesystem            Size  Used Avail Use% Mounted on/dev/mapper/vg_minimal-LogVol00                       18G  1.5G   15G   9% /tmpfs                 491M     0  491M   0% /dev/shm/dev/sda1             477M   61M  391M  14% /boot/dev/md0              4.0G  6.1M  3.8G   1% /mnt/raid

4.查看当前磁盘阵列信息

[root@minimal ~]# mdadm --detail /dev/md0/dev/md0:        Version : 1.2  Creation Time : Wed Oct 21 11:36:44 2015     Raid Level : raid5     Array Size : 3176448 (3.03 GiB 3.25 GB)  Used Dev Size : 1058816 (1034.17 MiB 1084.23 MB)   Raid Devices : 5  Total Devices : 6    Persistence : Superblock is persistent    Update Time : Wed Oct 21 12:31:14 2015          State : clean, reshaping  Active Devices : 5Working Devices : 6 Failed Devices : 0  Spare Devices : 1         Layout : left-symmetric     Chunk Size : 512K Reshape Status : 33% complete  Delta Devices : 1, (4->5)           Name : minimal:0  (local to host minimal)           UUID : f99d657f:0db77911:9870b07b:92c4c1d9         Events : 67    Number   Major   Minor   RaidDevice State       0       8       17        0      active sync   /dev/sdb1       1       8       18        1      active sync   /dev/sdb2       2       8       19        2      active sync   /dev/sdb3       4       8       22        3      active sync   /dev/sdb6       6       8       21        4      active sync   /dev/sdb5       5       8       23        -      spare   /dev/sdb7

或者也可以查阅mdstat文件来获取磁盘阵列的信息：

[root@minimal ~]# cat /proc/mdstatPersonalities : [raid6] [raid5] [raid4] md0 : active raid5 sdb5[6] sdb7[5](S) sdb6[4] sdb3[2] sdb2[1] sdb1[0]      4235264 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]      unused devices: <none>

四、关闭软件磁盘阵列

1.卸载磁盘并且删除与磁盘挂载的内容

[root@minimal ~]# umount /dev/md0

如果设置了磁盘整列开机自动挂载的话，还需要进入/etc/fstab删除自动挂载的内容

2.停止磁盘阵列服务

[root@minimal ~]# mdadm --stop /dev/md0mdadm: stopped /dev/md0

3.删除磁盘阵列配置文件

[root@minimal ~]# vim /etc/mdadm.confARRAY /dev/md0 metadata=1.2 spares=1 name=minimal:0 UUID=f99d657f:0db77911:9870b07b:92c4c1d9    #删除或者注释掉

0 0