[Galaxy Kirin V10] [server] soft RAID configuration

One 、 Experimental environment

Galaxy Unicorn V10  0711 server  x86_64

Two 、 With raid5 For example

raid5 yes raid0 and raid1 A compromise of , With and raid0 Similar data reading speed , Just one more parity message , The speed of writing data is slightly slower than that of a single disk , At the same time, because multiple data correspond to one parity information ,raid5 The disk space utilization ratio of raid1 high , Storage costs are relatively low , It's a solution that's used a lot .

Number of hard disks required ： At least 3 block , Can exceed 3 block ;

Hard disk utilization ：(n-1)/n  , among n Is the number of hard disks , Parity data actually adds up to one disk ;

Maximum disk failures ： Bad at most 1 Block plate ;

1、raid5 To configure

Prerequisite ： Get ready 4 Block hard disk , Here to sdb、sdc、sdd、sde For example , Self partition is sdb1、sdc1、sdd1、sde1

# mkfs.ext4   /dev/sdb1      // format partition

# mkfs.ext4  /dev/sdc1

# mkfs.ext4  /dev/sdd1

# mkfs.ext4  /dev/sde1

In the following ways 1 And way 2 Select an operation according to the actual needs ：

The way 1：3 Block plate , do raid5, No redundant disk

# mdadm --create --auto=yes /dev/md5 --level=5 --raid-devices=3 /dev/sdb1 /dev/sdc1 /dev/sdd1

Parameter description ：

--create： establish raid

--auto=yes： Decide to create a subsequent software disk array device , namely md[0-9]

/dev/md5：md5 by raid Device file name , Custom name , General habits raid5 Just name it md5, Easy to understand

--raid-devices=4： A device that uses several disks or partitions as a disk array

--level=5： Set the level of this group of disk arrays

The way 2：4 Block plate , do raid5, There is a redundant disk , If there is a fault during use , The redundant disk can automatically replace the bad disk

# mdadm --create /dev/md5 --auto=yes --level=5 --chunk=256K --raid-devices=3 --spare-devices=1 /dev/sdb1 /dev/sdc1 /dev/sdd1 /dev/sde1

Parameter description ：

--chunk=256K： Determine the of this device chunk size , It's usually 64K perhaps 512K

--spare-devices=1： Use several disks or partitions as backup devices

 # mdadm -D /dev/md5    // View the created raid Device details

[[email protected] ~]# mdadm -D /dev/md5
/dev/md5:     #raid Device file name 
           Version : 1.2
     Creation Time : Mon Nov  1 19:31:24 2021       # Creation time 
        Raid Level : raid5      #raid Grade 
        Array Size : 10473472 (9.99 GiB 10.72 GB)      # Whole set raid Actual usable capacity 
     Used Dev Size : 5236736 (4.99 GiB 5.36 GB)        # Capacity of each disk 
      Raid Devices : 3      # form raid Number of disks 
     Total Devices : 4      # Include spare Total disks 
       Persistence : Superblock is persistent

       Update Time : Mon Nov  1 19:31:50 2021
             State : clean      # Current usage status of this disk array 
    Active Devices : 3          # Number of devices started 
   Working Devices : 4          # The number of devices currently using this array 
    Failed Devices : 0          # Number of damaged equipment 
     Spare Devices : 1          # Number of reserved disks 

            Layout : left-symmetric
        Chunk Size : 256K       #chunk Small block capacity 

Consistency Policy : resync

              Name : 192.168.1.13:5  (local to host 192.168.1.13)
              UUID : c18e80e3:2786d97c:8fbdd2b1:c9b8c07b
            Events : 18

    Number   Major   Minor   RaidDevice State
       0       8       17        0      active sync   /dev/sdb1
       1       8       33        1      active sync   /dev/sdc1
       4       8       49        2      active sync   /dev/sdd1

       3       8       65        -      spare   /dev/sde1   # notice sde1 As standby equipment in the waiting area 

# Last 4 OK, here it is 4 Current status of devices ,RaidDevice  Refers to this Raid Disk order in 

# cat /proc/mdstat       // Check the status

[[email protected] ~]# cat /proc/mdstat 
Personalities : [raid6] [raid5] [raid4] 
md5 : active raid5 sdd1[4] sde1[3](S) sdc1[1] sdb1[0]
      10473472 blocks super 1.2 level 5, 256k chunk, algorithm 2 [3/3] [UUU]
      
unused devices: <none>

explain ： The first line is ： Pointed out that md05 by raid5 , And used sdd1,sde1,sdc1,sdb1, Wait for four disk devices . Brackets after each device [] The number in this disk is in RAID Order in （RaidDevice）; as for sde1  hinder [S] Then represent sde1  by spare The meaning of ;
The second line is ： This disk array has 10473472  individual block（ Every block Unit is 1K）, So the total capacity is about 9.98GB, Use RAID 5 Grade , Small block written to disk （chunk） The size is 256K, Use algorithm 2 Disk array algorithm . [m/n] Represents that this array requires m Devices , And n One device is working properly . So this md5  need 3  A device and this 3 All devices are in normal operation . hinder [UUU] It represents four required equipment （ Namely [m/n] Inside m） Start up of ,U Represents normal operation , if _ It means abnormal .

# mkfs.ext4 /dev/md5     // format

# blkid  /dev/md5    // obtain UUID, write in fstab Realize automatic mount after power on

 # mkdir   /data2    // Create mount directory , Name and path customization

# vim  /etc/fstab

 # mount  -a     // If you enter and no error is reported, the mount is successful , This step is very important , Can be tested fstab Whether the writing is correct

# df  -Th     //raid Disk mounted successfully