Linux☞How to implement soft RAID 0, soft RAID 1, and soft RAID 10

Raid implementation:
Raid: cheap redundant disk array, which turns several hard disks into a whole
Later, because it lost its original cheap purpose and became very expensive, it changed its name
Independent redundant disk array
Although we cannot implement raid on hardware, we can still implement raid because of our limited strength and we didn't buy those devices.
Next we will look at how to implement soft RAID:

1. Implementation of soft RAID 0:
First, we will introduce the principle of RAID 0. RAID 0 divides a data stream into two copies and stores them on two disks to achieve doubled performance.
Not only that, but it also provides a large throughput. Next we will implement soft RAID 0.
First, you must create two partitions (RAID 0 can be created if there are more than two partitions). When creating a raid with a size of 2 GB, you must
The two shards are 1 GB in size. After the shards are created, repeat the partition table in the kernel,

partprobe /dev/sda[root@server56 raid10]# fdisk /dev/sdaCommand (m for help): nFirst cylinder (7011-15665, default 7011):      Using default value 7011Last cylinder or +size or +sizeM or +sizeK (7011-15665, default 15665): +1GCommand (m for help): tPartition number (1-6): 5Hex code (type L to list codes): fdChanged system type of partition 5 to fd (Linux raid autodetect)Command (m for help): tPartition number (1-6): 6Hex code (type L to list codes): fdChanged system type of partition 6 to fd (Linux raid autodetect)   Device Boot      Start         End      Blocks   Id  System/dev/sda1   *           1          13      104391   83  Linux/dev/sda2              14        5235    41945715   8e  Linux LVM/dev/sda3            5236        5300      522112+  82  Linux swap / Solaris/dev/sda4            5301       15665    83256862+   5  Extended/dev/sda5            5301        5423      987966   fd  Linux raid autodetect/dev/sda6            5424        5546      987966   fd  Linux raid autodetect

In Linux, raid autodetect is used to establish soft RAID. The system type is required. Remember to divide the two raid files when partitioning.
The partition is specified to this type. The following lists all the numbers of the system type and system type that can be specified for the partition.
You only need to enter the preceding number.

0  Empty           1e  Hidden W95 FAT1 80  Old Minix       bf  Solaris         1  FAT12           24  NEC DOS         81  Minix / old Lin c1  DRDOS/sec (FAT- 2  XENIX root      39  Plan 9          82  Linux swap / So c4  DRDOS/sec (FAT- 3  XENIX usr       3c  PartitionMagic  83  Linux           c6  DRDOS/sec (FAT- 4  FAT16 <32M      40  Venix 80286     84  OS/2 hidden C:  c7  Syrinx          5  Extended        41  PPC PReP Boot   85  Linux extended  da  Non-FS data     6  FAT16           42  SFS             86  NTFS volume set db  CP/M / CTOS / . 7  HPFS/NTFS       4d  QNX4.x          87  NTFS volume set de  Dell Utility    8  AIX             4e  QNX4.x 2nd part 88  Linux plaintext df  BootIt          9  AIX bootable    4f  QNX4.x 3rd part 8e  Linux LVM       e1  DOS access      a  OS/2 Boot Manag 50  OnTrack DM      93  Amoeba          e3  DOS R/O         b  W95 FAT32       51  OnTrack DM6 Aux 94  Amoeba BBT      e4  SpeedStor       c  W95 FAT32 (LBA) 52  CP/M            9f  BSD/OS          eb  BeOS fs         e  W95 FAT16 (LBA) 53  OnTrack DM6 Aux a0  IBM Thinkpad hi ee  EFI GPT         f  W95 Ext'd (LBA) 54  OnTrackDM6      a5  FreeBSD         ef  EFI (FAT-12/16/10  OPUS            55  EZ-Drive        a6  OpenBSD         f0  Linux/PA-RISC b11  Hidden FAT12    56  Golden Bow      a7  NeXTSTEP        f1  SpeedStor      12  Compaq diagnost 5c  Priam Edisk     a8  Darwin UFS      f4  SpeedStor      14  Hidden FAT16 <3 61  SpeedStor       a9  NetBSD          f2  DOS secondary  16  Hidden FAT16    63  GNU HURD or Sys ab  Darwin boot     fb  VMware VMFS    17  Hidden HPFS/NTF 64  Novell Netware  b7  BSDI fs         fc  VMware VMKCORE 18  AST SmartSleep  65  Novell Netware  b8  BSDI swap       fd  Linux raid auto1b  Hidden W95 FAT3 70  DiskSecure Mult bb  Boot Wizard hid fe  LANstep        1c  Hidden W95 FAT3 75  PC/IX           be  Solaris boot    ff  BBT  

After the partition is created, enter W to save and exit. If Q is used to exit, all the partitions above will be unavailable.

After the partition is created, remember to format it and create the file system ext2 or ext3 for it. No matter what type of file system you specify
The file system type of the partition must be the same to ensure consistency and synchronization.

mkfs.ext2 /dev/sda5mkfs.ext2 /dev/sda6

After formatting is complete, you can proceed to the next step and create a soft RAID 0

mdadm -C /dev/md0 -a yes -l 0 -n 2 /dev/sda5 /dev/sda6

[This command is used to create/dev/md0. If not, create/dev/md0 directly. The raid level is 0 and
/Dev/SD6 components]

After the partition is created, the entire partition is formatted because the two partitions belong to one hard disk logically,

mke2fs /dev/md0

Now you can take a look at/proc/mdstat

md0 : active raid0 sda6[1] sda5[0]      1975680 blocks 64k chunks

The Chunks here is the minimum data unit for disk read/write. The unit is K. The default value is 64 K.
The default data block size is 4096 kb. Therefore, each time the kernel calculates the number of chunks read
However

mk2tfs -b 2048 -E stride=32 /dev/md0

You can directly tell the kernel the number of chunks to be read, which can greatly
Saves CPU space.-B is the size of a data block.

Next, you need to specify a mount point for RAID 0.

mkdir /mnt/raid0mount /dev/nd0 /mnt/raid0

You can mount and use it,
Use fdisk-L to view

Mdadm-D/dev/md0 to View Details

In this case, you should save the configuration information of RAID 0 in case of a fault. You can recover it next time,

mdadm -D --scan /dev/md0 >> /etc/mdadm.conf

You can use it directly next time.

Mdadm-A -- Scan direct mounting

How to detach RAID 0 should first detach the mount point of RAID 0, that is

umount /mnt/raid0

Then use the command

mdadm -S /dev/md0 

Stop it
In this case, you can use fdisk-L to unmount it.
Directly

mdadm -A --scan

The kernel will automatically read/etc/mdadm. conf to mount all the soft RAID

In case of an accident, if a disk with RAID 0 breaks down, all the data will be lost, so RAID 1 appears.
First, we will introduce the principle of RAID 1. Raid 1 stores data streams on two disks at the same time, just like a mirror.
Data, even if one of them is unfortunately down, there is still complete data, but the two disks have the same data
The space of one disk is wasted, but it still provides powerful read capability, although the write speed will be the same as the original
Living is not as good as before, but there is still a strong Throughput
To create a RAID 1 with a size of 2 GB, it takes so long for two disks with the same size as 2 GB, that is, two partitions with the same size as 2 GB.
Implement soft RAID 1
First, two partitions with a size of 2 GB are required. The system type is still
Linux raid autodetect
Repeat the partition

partprobe /dev/sda

Partition and format

mkfs.ext3 /dev/sda7 mkfs.ext3 /dev/sda8

Then you can create soft RAID 1.

mdadm -C /dev/md1 -a yes -l 1 -n 2 /dev/sda7 /dev/sda8

Then, format the entire file.
Mkfs. ext3 note that the format must be consistent
Create a mount point

mkdir /mnt/raid1

Mount raid1

mount /dev/md1 /mnt/raid1

Now you can use
You can check the LS-L in CD, MNT, and raid1.

drwx------ 2 root root 16384 07-30 17:00 lost+found

You can also see fdisk-l

Remember to save the relevant information.

mdadm -D --scan /dev/md1 >> /etc/mdadm.conf

The steps for detaching a RAID 0 are the same as those for detaching a RAID 0,
You only need

Mdadm-A -- scan to mount all RAID Devices

Even if an unexpected raid occurs, it can work normally, but it is working in a lame State. At this time, it is easy to cause another disk.
If the disk is damaged, you need to add a backup disk. The same is true for Soft Raid. A backup partition is required for reserve.
Then, the system type is still FD, which is the same as the first two shards.
After the partition is complete, re-read the partition table and format it into the same file system as RAID 1. Then
We will simulate a partition failure and use the command

Mdadm -- fail/dev/md1/dev/sda7. The sda7 in md1 is crashed.

Now we need to use the partition just now,
Install a good disk and then a bad disk to avoid secondary harm.

Mdadm -- add/dev/md1/dev/sda9 [after installation, it will automatically synchronize with another disk]

Remove a bad Disk:

mdadm --remove /dev/md1 /dev/sda7

Because RAID 0 and RAID 1 both have shortcomings and all have certain risks, the combination of the two methods will reduce the number
Risks, and provides excellent read/write performance. with powerful throughput, RAID 0 is used as the middle layer to improve read/write performance.
Capability. Raid 1 is used as the underlying layer to provide Fault Tolerance capabilities, which reduces the risk of data loss.

To implement soft RAID 10, ensure that RAID 0 and RAID 1 are not mounted and are in the unmounted status.

mdadm -C /dev/md10 -a yes -l 10 -n 2 /dev/md0 /dev/md1

In this way, soft RAID 10 can be implemented.
No formatting is required.
You can mount it directly.
Create a mount point first

mkdir /mnt/raid10

Mount

mount/dev/md10 /mnt/raid10

That is, soft RAID 10 is implemented.

In this case, save the configuration information

mdadm -D /dev/md10 >> /etc/mdadm.conf

The uninstall method is the same as the first two methods.
Reload is the same as before

Mdadm-A -- scan can

In this way, the above three types of soft RAID are implemented.

Raid is a topic for interviews, so you need to be familiar with it !!!
Next, let's try RAID 5.
The method is similar.