Basic work with LVM in Linux

To start - check the partitions:

# cat /proc/partitions
major minor  #blocks  name
  11        0    1920096 sr0
 252        0   20971520 vda
 252        1    1048576 vda1
 252        2   19921920 vda2
   8       16    1048576 sdb
   8        0    1048576 sda
 253        0   17821696 dm-0
 253        1    2097152 dm-1

I have sda and sdb: both are 1 GB in size.

So, first create a Volume Group named vgstriped:

# vgcreate vgstriped /dev/sda /dev/sdb
  Physical volume "/dev/sda" successfully created.
  Physical volume "/dev/sdb" successfully created.
  Volume group "vgstriped" successfully created

Show what we have:

# pvs
  PV         VG                    Fmt  Attr PSize    PFree   
  /dev/sda   vgstriped             lvm2 a--  1020.00m 1020.00m
  /dev/sdb   vgstriped             lvm2 a--  1020.00m 1020.00m
  /dev/vda2  fedora_localhost-live lvm2 a--   <19.00g       0

Let's use lvcreate to create a striped logical volume named lvdata:

# lvcreate -L 1G -i2 -I64 -n lvdata vgstriped 
  Logical volume "lvdata" created.

Options:

-L 1G - the size of logical volume
-i2 - amount of volumes that is should be striped on
-I64 - will define a stripe size of 64 KB
-n lvdata - volume name will be lvdata

Let's use lvs to see if it was successful:

# lvs
  LV     VG                    Attr       LSize   Pool Origin Data%  Meta%  Move Log Cpy%Sync Convert
  root   fedora_localhost-live -wi-ao---- <17.00g                                                    
  swap   fedora_localhost-live -wi-ao----   2.00g                                                    
  lvdata vgstriped             -wi-a-----   1.00g

If you want more information use lvdisplay command:

# lvdisplay /dev/vgstriped/lvdata 
  --- Logical volume ---
  LV Path                /dev/vgstriped/lvdata
  LV Name                lvdata
  VG Name                vgstriped
  LV UUID                4beSCk-Zffc-dvRc-dRBm-zYex-ref6-ra06ZH
  LV Write Access        read/write
  LV Creation host, time localhost.localdomain, 2021-12-24 21:44:06 +0200
  LV Status              available
  # open                 0
  LV Size                1.00 GiB
  Current LE             256
  Segments               1
  Allocation             inherit
  Read ahead sectors     auto
  - currently set to     512
  Block device           253:2

likewise there are vgdisplay and pvdisplay commands.

Once we created a logical volume, let's put a filesystem an top of it:

# mkfs.ext4 /dev/vgstriped/lvdata 
mke2fs 1.45.5 (07-Jan-2020)
Discarding device blocks: done                            
Creating filesystem with 262144 4k blocks and 65536 inodes
Filesystem UUID: 4f97cdad-a383-4b34-9e92-a6ca7d17d570
Superblock backups stored on blocks: 
	32768, 98304, 163840, 229376

Allocating group tables: done                            
Writing inode tables: done                            
Creating journal (8192 blocks): done
Writing superblocks and filesystem accounting information: done

And now we have a filesystem, we can mount it:

# mount /dev/vgstriped/lvdata /mnt/
# df -h
Filesystem                               Size  Used Avail Use% Mounted on
devtmpfs                                 970M     0  970M   0% /dev
tmpfs                                    990M     0  990M   0% /dev/shm
tmpfs                                    990M  1.5M  989M   1% /run
/dev/mapper/fedora_localhost--live-root   17G  7.2G  8.6G  46% /
tmpfs                                    990M  100K  990M   1% /tmp
/dev/vda1                                976M  179M  731M  20% /boot
tmpfs                                    198M   96K  198M   1% /run/user/1000
/dev/sr0                                 1.9G  1.9G     0 100% /run/media/user/Fedora-WS-Live-32-1-6
/dev/mapper/vgstriped-lvdata             976M  2.6M  907M   1% /mnt

Let's copy some data to this mounted volume:

# cp -r /etc/[abc]* /mnt/

Now, we have some data in there:

# cd /mnt/
# ls
abrt     alternatives  at.deny     bash_completion.d       bluetooth    ceph         chrony.keys  cron.d       cron.monthly     crypttab   cupshelpers
adjtime  anaconda      audit       bashrc                  brlapi.key   chkconfig.d  cifs-utils   cron.daily   crontab          csh.cshrc  lost+found
aliases  anacrontab    authselect  bindresvport.blacklist  brltty       chromium     cni          cron.deny    cron.weekly      csh.login
alsa     asound.conf   avahi       binfmt.d                brltty.conf  chrony.conf  containers   cron.hourly  crypto-policies  cups

One of benefits of working with LVM is that you can easily resize LVM volumes. If we type vgs:

# vgs
  VG                    #PV #LV #SN Attr   VSize   VFree   
  fedora_localhost-live   1   2   0 wz--n- <19.00g       0 
  vgstriped               2   1   0 wz--n-   1.99g 1016.00m

we can see that we have VFree of approximately 1G in vgstriped.
If you have available disk space in the Volume Group it is very easy to resize, just add more disk space.
If you do not have available disk space in the Volume Group - you need to add additional physical volumes, otherwise you cannot resize.

There are a couple of options that can be used in lvresize command:

# lvresize --help
  lvresize - Resize a logical volume

  Resize an LV by a specified size.
  lvresize -L|--size [+|-]Size[m|UNIT] LV
	[ -l|--extents [+|-]Number[PERCENT] ]
	[ -r|--resizefs ]
	[    --poolmetadatasize [+]Size[m|UNIT] ]
	[ COMMON_OPTIONS ]
	[ PV ... ]

  Resize an LV by specified PV extents.
  lvresize LV PV ...
	[ -r|--resizefs ]
	[ COMMON_OPTIONS ]

  Resize a pool metadata SubLV by a specified size.
  lvresize --poolmetadatasize [+]Size[m|UNIT] LV_thinpool
	[ COMMON_OPTIONS ]
	[ PV ... ]

  Common options for command:
	[ -A|--autobackup y|n ]
	[ -f|--force ]
	[ -i|--stripes Number ]
	[ -I|--stripesize Size[k|UNIT] ]
	[ -n|--nofsck ]
	[    --alloc contiguous|cling|cling_by_tags|normal|anywhere|inherit ]
	[    --nosync ]
	[    --noudevsync ]
	[    --reportformat basic|json ]
	[    --type linear|striped|snapshot|mirror|raid|thin|cache|vdo|thin-pool|cache-pool|vdo-pool ]

  Common options for lvm:
	[ -d|--debug ]
	[ -h|--help ]
	[ -q|--quiet ]
	[ -v|--verbose ]
	[ -y|--yes ]
	[ -t|--test ]
	[    --commandprofile String ]
	[    --config String ]
	[    --driverloaded y|n ]
	[    --nolocking ]
	[    --lockopt String ]
	[    --longhelp ]
	[    --profile String ]
	[    --version ]

  Use --longhelp to show all options and advanced commands.

options -l and -L allows you to specify the size that should be added. -l allows to specify the percentage of free extets, With -L you can use absolute number +200M for example.
-r - is very interesting option which should always be used, because it will automatically resize the filesystem as well

So, let's do a lvresize:

# lvresize -L +100M -r /dev/vgstriped/lvdata
  Using stripesize of last segment 64.00 KiB
  Rounding size (281 extents) up to stripe boundary size for segment (282 extents).
  Size of logical volume vgstriped/lvdata changed from 1.00 GiB (256 extents) to 1.10 GiB (282 extents).
  Logical volume vgstriped/lvdata successfully resized.
resize2fs 1.45.5 (07-Jan-2020)
Filesystem at /dev/mapper/vgstriped-lvdata is mounted on /mnt; on-line resizing required
old_desc_blocks = 1, new_desc_blocks = 1
The filesystem on /dev/mapper/vgstriped-lvdata is now 288768 (4k) blocks long.

Options used:

-L +100M - to add 100Mb
-r - to resize the filesystem

We can now see the size of the filesystem is set to 1.1Gb:

# df -h
Filesystem                               Size  Used Avail Use% Mounted on
devtmpfs                                 970M     0  970M   0% /dev
tmpfs                                    990M     0  990M   0% /dev/shm
tmpfs                                    990M  1.5M  989M   1% /run
/dev/mapper/fedora_localhost--live-root   17G  7.2G  8.6G  46% /
tmpfs                                    990M  100K  990M   1% /tmp
/dev/vda1                                976M  179M  731M  20% /boot
tmpfs                                    198M   96K  198M   1% /run/user/1000
/dev/sr0                                 1.9G  1.9G     0 100% /run/media/user/Fedora-WS-Live-32-1-6
/dev/mapper/vgstriped-lvdata             1.1G  7.1M 1000M   1% /mnt

How to create the snapshot.
A snapshot allows you to freeze the state of the logical volume which means that everything that will be added/removed after you created the snapshot, won't be visible in the snapshot.
So, snapshots can be an easy way to revert to a previous state and snapshots can also help you to create backups on a filesystem where no files are open.

To create a snapshot:

# lvcreate -s -n lvdata_snap -L 100M /dev/vgstriped/lvdata
  Logical volume "lvdata_snap" created.

Options:

-s - create snapshot
-n lvdata_snap - the name of snapshot
-L 100M - the size of snapshot

Show the snapshot details:

# lvs
  LV          VG                    Attr       LSize   Pool Origin Data%  Meta%  Move Log Cpy%Sync Convert
  root        fedora_localhost-live -wi-ao---- <17.00g                                                    
  swap        fedora_localhost-live -wi-ao----   2.00g                                                    
  lvdata      vgstriped             owi-aos---   1.10g                                                    
  lvdata_snap vgstriped             swi-a-s--- 100.00m      lvdata 0.01

you can see that lvdata_snap with origin lvdata has been created.

What is good about snapshots? One of the things is that they keep the original state of the logical volume.
The logical volume is still mounted on /mnt, so let's remove files starting with b:

# cd /mnt/
# rm -rf b*

Now, let's unmount /mnt and mount the snapshot on /mnt:

# cd
# umount /mnt 
# mount /dev/vgstriped/lvdata_snap /mnt/

And we will see that the files starting with b are still exists:

# cd /mnt/
# ls
abrt     alternatives  at.deny     bash_completion.d       bluetooth    ceph         chrony.keys  cron.d       cron.monthly     crypttab   cupshelpers
adjtime  anaconda      audit       bashrc                  brlapi.key   chkconfig.d  cifs-utils   cron.daily   crontab          csh.cshrc  lost+found
aliases  anacrontab    authselect  bindresvport.blacklist  brltty       chromium     cni          cron.deny    cron.weekly      csh.login
alsa     asound.conf   avahi       binfmt.d                brltty.conf  chrony.conf  containers   cron.hourly  crypto-policies  cups

That is what happening in the snapshot. Now, in the background LVM is working on copying all deleted files from the original Logical Volume to the snapshot.
So, with every deletion. on the original Logical Volume - your snapshot will grow bigger. Which is something that could be monitored as well:

# lvs
  LV          VG                    Attr       LSize   Pool Origin Data%  Meta%  Move Log Cpy%Sync Convert
  root        fedora_localhost-live -wi-ao---- <17.00g                                                    
  swap        fedora_localhost-live -wi-ao----   2.00g                                                    
  lvdata      vgstriped             owi-a-s---   1.10g                                                    
  lvdata_snap vgstriped             swi-aos--- 100.00m      lvdata 0.33

Previously the Data% was set to 0.01 and you can now see that it is set to something a lot higher.
Note that if your original Logical Volume goes away, then the snapshot becomes useless as well.