>FOR FREEDOM!<
{A} Introduction
Here's a short description of what is supported in the Linux RAID drivers. RAID is not a guarantee for data integrity, it just allows you to keep your data if a disk dies.
The current RAID drivers in Linux support the following levels:
Linear mode | RAID-0 | RAID-1
RAID-4
- If one drive fails, the parity information can be used to reconstruct all data. If two drives fail, all data is lost.
- The reason this level is not more frequently used, is because the parity information is kept on one drive. This information must be updated every time one of the other disks are written to. Thus, the parity disk will become a bottleneck, if it is not a lot faster than the other disks. However, if you just happen to have a lot of slow disks and a very fast one, this RAID level can be very useful.
RAID-5
- This is perhaps the most useful RAID mode when one wishes to combine a larger number of physical disks, and still maintain some redundancy. RAID-5 can be (usefully) used on three or more disks, with zero or more spare-disks. The resulting RAID-5 device size will be (N-1)*S, just like RAID-4. The big difference between RAID-5 and -4 is, that the parity information is distributed evenly among the participating drives, avoiding the bottleneck problem in RAID-4, and also getting more performance out of the disk when reading, as all drives will then be used.
- If one of the disks fail, all data are still intact, thanks to the parity information. If spare disks are available, reconstruction will begin immediately after the device failure. If two disks fail simultaneously, or before the raid is reconstructed, all data are lost. RAID-5 can survive one disk failure, but not two or more.
- Both read and write performance usually increase, but can be hard to predict how much. Reads are almost similar to RAID-0 reads, writes can be either rather expensive (requiring read-in prior to write, in order to be able to calculate the correct parity information, such as in database operations), or similar to RAID-1 writes (when larger sequential writes are performed, and parity can be calculated directly from the other blocks to be written). The write efficiency depends heavily on the amount of memory in the machine, and the usage pattern of the array. Heavily scattered writes are bound to be more expensive.
RAID-6
- This is an extension of RAID-5 to provide more resilience. RAID-6 can be (usefully) used on four or more disks, with zero or more spare-disks. The resulting RAID-6 device size will be (N-2)*S. The big difference between RAID-5 and -6 is that there are two different parity information blocks, and these are distributed evenly among the participating drives.
- Since there are two parity blocks; if one or two of the disks fail, all data is still intact. If spare disks are available, reconstruction will begin immediately after the device failure(s).
- Read performance is almost similar to RAID-5 but write performance is worse.
RAID-10
- RAID-10 is an "in-kernel" combination of RAID-1 and RAID-0 that is more efficient than simply layering RAID levels.
- RAID-10 has a layout ("far") which can provide sequential read throughput that scales by number of drives, rather than number of RAID-1 pairs. You can get about 95 % of the performance of the RAID-0 with same amount of drives.
- RAID-10 allows spare disk(s) to be shared amongst all the raid1 pairs.
FAULTY
- This is a special debugging RAID level. It only allows one device and simulates low level read/write failures.
- Using a FAULTY device in another RAID level allows administrators to practice dealing with things like sector-failures as opposed to whole drive failures
{B} Swapping on RAID
Swapping on a mirrored RAID can help you survive a failing disk. If a disk fails, then data for swapped processes would be inaccessable in a non-mirrored environment. If you run in a mirrored environment, then the system can go on running even if a disk fails in service.
There's not much reason to use RAID0 for swap performance reasons. The kernel itself can stripe swapping on several devices, if you just give them the same priority in the /etc/fstab file.
A nice /etc/fstab could look like:
/dev/sda2 none swap defaults,pri=4 0 0
/dev/sdb2 none swap defaults,pri=4 0 0
/dev/sdc2 none swap defaults,pri=4 0 0
/dev/sdd2 none swap defaults,pri=4 0 0
/dev/sde2 none swap defaults,pri=4 0 0
/dev/sdf2 none swap defaults,pri=4 0 0
/dev/sdg2 none swap defaults,pri=4 0 0
This setup lets the machine swap in parallel on seven SAS devices. No need for RAID0, since this has been a kernel feature for a long time.
A different reason to use RAID for swap is high availability. If you set up a system to boot on eg. a RAID-1 device, the system should be able to survive a disk crash. If a system without mirrored swapping has been swapping on the now faulty device, you will most likely be going down. Swapping on a mirrored RAID partition such as RAID-1, raid10,n2 or raid10,f2 type would solve this problem.
{C} Spare disks
Spare disks (often called hot spares) are disks that do not take part in the RAID set until one of the active disks fail. When a device failure is detected, that device is marked as "faulty" and reconstruction is immediately started on the first spare disk available.
once reconstruction to a hot-spare begins, the RAID layer will start reading from all the other disks to re-create the redundant information. If multiple disks have built up bad blocks over time, the reconstruction itself can actually trigger a failure on one of the "good" disks. This can lead to a complete RAID failure and is the major reason for using RAID-6 in preference to RAID-5 and a hot spare.
{D} Faulty disks
When the RAID layer handles device failures just fine, crashed disks are marked as faulty, and reconstruction is immediately started on the first spare-disk available. If no spare is available then the array runs in 'degraded' mode.
Faulty disks still appear and behave as members of the array. The RAID layer just avoids reading/writing them.
If a device needs to be removed from an array for any reason (eg pro-active replacement due to SMART reports) then it must be marked as faulty before it can be removed.
{E} RAID setup
Prepare
Install the package "mdadm", and "modprobe raid456"、“modprobe raid10” etc.Then you will see：
[root@6 ~]# cat /proc/mdstat Personalities : [raid10] [raid6] [raid5] [raid4] unused devices: <none>
Mdadm modes of operation
mdadm has 7 major modes of operation. Normal operation just uses the 'Create', 'Assemble' and 'Monitor' commands - the rest is typically used for fixing or changing your array.
- Create:Create a new array with per-device superblocks (normal creation).
- Assemble:Assemble the parts of a previously created array into an active array.
- Follow or Monitor:Monitor one or more md devices and act on any state changes.
- Build:Build an array that doesn't have per-device superblocks. [Rarely used！]
- Grow:Grow, shrink or otherwise reshape an array in some way. [Rarely used!]
- Manage:This is for doing things to specific components of an array such as adding new spares and removing faulty devices.
- Misc:This is an 'everything else' mode that supports operations on active arrays, operations on component devices such as erasing old superblocks, and information gathering operations.
Create the Partition Table (GPT)
It is highly recommended to pre-partition the disks to be used in the array.
Note: It is also possible to create a RAID directly on the raw disks (without partitions), but not recommended because it can cause problems when swapping a failed disk.
parted -a optimal /dev/vdX -mklabel gpt
parted -a optimal /dev/vdX mkpart 1M xM #x = total_Mb - 100M parted -a optimal /dev/vdX set 1 raid ... parted -a optimal /dev/vdZ -mklabel gpt
parted -a optimal /dev/vdZ mkpart 1M xM #x is the previous x, do not recalculate! parted -a optimal /dev/vdZ set 1 raid
Create RAID device
Raid0 mdadm --create --auto=mdp /dev/mdX --level=0 --raid-devices=26 /dev/vd{a..z}1
#If --auto is not given on the command line or in the config file, then the default will be --auto=yes
#"part" or "mdp" causes a partitionable array (2.6 and later) to be used
Raid1
mdadm --create /dev/mdX --level=1 --raid-devices=2 /dev/vd{a,b}1 --spare-devices=2 /dev/vd{c,d}1
Raid6
mdadm --create /dev/mdX --level=6 --raid-devices=4 /dev/vd{a..d}1 --spare-devices=1 /dev/vde1
Raid10 #Raid10 with “--layout=f2" algorithm perform best in reading data
mdadm --create --verbose /dev/mdX --metadata=1.2 --chunk=256 --level=10 --raid-devices=6 --layout=f2 /dev/vd{a..f}1 --spare-devices=2 /dev/vd{g,h}1
Remember to this for possiable assembling in the future:
# echo 'DEVICE partitions' > /etc/mdadm.conf # mdadm --detail --scan >> /etc/mdadm.conf
This results in something like the following:
root # cat /etc/mdadm.conf
DEVICE partitions ARRAY /dev/md/0 metadata=1.2 name=pine:0 UUID=27664f0d:111e493d:4d810213:9f291abe
Create partitions on array (or use LVM upon it,will discussing in the {H} chapter)
Same as normal disk-partitions: use parted OR gdisk And format them: mke2fs -t ext4 -b 4096 /dev/md0_pX ...
Removing devices from an array
- Mark it as faulty
-
mdadm --fail /dev/md0 /dev/sdxx
- Remove it from the array
-
mdadm -r /dev/md0 /dev/sdxx
- Remove device permanently(After the two commands described above)
-
mdadm --zero-superblock /dev/sdxx OR dd if=/dev/null of=/dev/sdxx bs=1M count=10
Warning: Reusing the removed disk without zeroing the superblock WILL CAUSE LOSS OF ALL DATA on the next boot. (After mdadm will try to use it as the part of the raid array).
Stop using an array
- Umount target array
- Stop the array with:
mdadm --stop /dev/md0 - Do "mdadm --zero-superblock /dev/vdxx" on each device
- Remove the corresponding line from /etc/mdadm.conf
Adding a New Device to an Array for repair or spare purpose(Not mean growing numbers of array!)
Adding new devices with mdadm can be done on a running system with the devices mounted. Partition the new device using the same layout as others in the same array.
- Assemble the RAID array if it is not already assembled
-
mdadm --assemble /dev/md0 /dev/sda1 /dev/sdb1
OR
mdadm --assemble UUID=27664f0d:111e493d:4d810213:9f291abe #Need "mdadm.conf" which must be prepared in adance - Add the new device the array
-
mdadm --add /dev/md0 /dev/sdc1
Change sync speed limits
Syncing can take a while. If the machine is not needed for other tasks the speed limit can be increased.
# cat /proc/mdstat Personalities : [raid1] md0 : active raid1 sda3[2] sdb3[1] 155042219 blocks super 1.2 [2/1] [_U] [>....................] recovery = 0.0% (77696/155042219) finish=265.8min speed=9712K/sec unused devices: <none>
Check the current speed limit.
# cat /proc/sys/dev/raid/speed_limit_min 1000 # cat /proc/sys/dev/raid/speed_limit_max 200000
Increase the limits.
# echo 400000 >/proc/sys/dev/raid/speed_limit_min # echo 400000 >/proc/sys/dev/raid/speed_limit_max
Then check out the syncing speed and estimated finish time
# cat /proc/mdstat Personalities : [raid1] md0 : active raid1 sda3[2] sdb3[1] 155042219 blocks super 1.2 [2/1] [_U] [>....................] recovery = 1.3% (2136640/155042219) finish=158.2min speed=16102K/sec unused devices: <none>
{F} Further reading
Calculating the Stride and Stripe-width
The array will have an entry in
# /sys/devices/virtual/block/mdX/queue/optimal_io_size
(where mdX is the name of your array). It will give the stripe-width in bytes. Divide by the block size to get the stripe width in blocks, then divide by number of data disks to get the stride. The following calculations should match this.
Stride = (chunk size/block size)
what is a reasonable chunk size?
- It depends on your average I/O request size. Here's the rule of thumb: big I/Os = small chunks; small I/Os = big chunks.
- Tip: See also Chunks: the hidden key to RAID performance.
Next, calculate:
Stripe-width = (# of physical data disks * stride)
Example: RAID10,far2[formatting to ext4 with the correct stripe-width and stride]
# cat /sys/devices/virtual/block/md0/queue/optimal_io_size # 1048576
Hypothetical RAID10 array is composed of 2 physical disks. Because of the properties of RAID10 in far2 layout, both count as data disks. Chunk size is 512k. Block size is 4k. So the stripe-width should match 1048576 / 4096 = 256, and the stride should match 256 / 2 = 128. Stride = (chunk size/block size). In this example, the math is (512/4) so the stride = 128. Stripe-width = (# of physical data disks * stride). In this example, the math is (2*128) so the stripe-width = 256. # mkfs.ext4 -v -L myarray -m 0.01 -b 4096 -E stride=128,stripe-width=256 /dev/md0
{G} How to replace the broken disks?
Remove all usage of the failed disk
- mdadm --manage /dev/mdX --remove /dev/sdX
- umount /dev/sdX*
(FIRST) Remove the data cable of the failed disk
(SECOND) Remove the power cable of the failed disk
- Force system to re-scan
- echo "- - -" > /sys/class/scsi_host/hostX/scan # For all "X"
- tail -f /var/log/syslog OR journalctl -kf # is a good idea
Replace the failed disk
(FIRST) Connect the power cable of the new disk (and wait some seconds)
(SECOND) Connect the data cable of the new disk
- Force system to re-scan
- echo "- - -" > /sys/class/scsi_host/hostX/scan # For all "X"
- tail -f /var/log/syslog OR journalctl -kf # is a good idea
{H} Linux LVM
- pvcreate vgcreate lvcreate
- pvmove
- pvremove vgremove lvremove
- pvscan vgscan lvsan
- pvdispaly vgdisplay lvdisplay
- vgreduce lvreduce
- vgextend lvextend
If a physical volume needs to be removed from a volume group, the data first needs to be moved away from the physical volume. With the pvmove command, all data on a physical volume is moved to other physical volumes within the same volume group.
root #pvmove -v /dev/sda1
Such an operation can take a while depending on the amount of data that needs to be moved. Once finished, there should be no data left on the device. Verify with pvdisplay that the physical volume is no longer used by any logical volume.
If a logical volume needs to be reduced in size, first shrink the file system itself. Not all file systems support online shrinking.For instance, ext4 does not support online shrinking so the file system needs to be unmounted first. It is also recommended to do a file system check to make sure there are no inconsistencies:
root #umount /mnt/data
root #e2fsck -f /dev/vg0/lvol1
root #resize2fs /dev/vg0/lvol1 150M
root #lvreduce --size 150M /dev/vg0/lv0l1
An extended volume group does not immediately provide the additional storage to the end users. For that, the file system on top of the volume group needs to be increased in size as well. Not all file systems allow online resizing!
For instance, to resize an ext4 file system to become 500MB in size:
lvextend --size 500M /dev/vg0/lv0l1
resize2fs /dev/vg0/lvol1 500M
OR combine two steps in one:
lvextend --resizefs --size 500M /dev/vg0/lv0l1
Create snapshot:
lvcreate --size 1G --snapshot --name lv0-snapshot --permission r[w] /dev/vg0/lv0
REFERENCE
- https://wiki.archlinux.org/index.php/RAID
- https://raid.wiki.kernel.org/index.php/Linux_Raid
- https://wiki.gentoo.org/wiki/LVM
- https://wiki.archlinux.org/index.php/LVM