How to Take ‘Snapshot of Logical Volume and Restore’ in LVM – Part III

LVM Snapshots are space efficient pointing time copies of lvm volumes. It works only with lvm and consume the space only when changes are made to the source logical volume to snapshot volume. If source volume has a huge changes made to sum of 1GB the same changes will be made to the snapshot volume. Its best to always have a small size of changes for space efficient. Incase the snapshot runs out of storage, we can use lvextend to grow. And if we need to shrink the snapshot we can use lvreduce.

Take Snapshot in LVM

Take Snapshot in LVM

If we have accidentally deleted any file after creating a Snapshot we don’t have to worry because the snapshot have the original file which we have deleted. It is possible if the file was there when the snapshot was created. Don’t alter the snapshot volume, keep as it while snapshot used to do a fast recovery.

Snapshots can’t be use for backup option. Backups are Primary Copy of some data’s, so we cant use snapshot as a backup option.

Requirements

  1. Create Disk Storage with LVM in Linux – PART 1
  2. How to Extend/Reduce LVM’s in Linux – Part II
My Server Setup
  1. Operating System – CentOS 6.5 with LVM Installation
  2. Server IP – 192.168.0.200

Step 1: Creating LVM Snapshot

First, check for free space in volume group to create a new snapshot using following ‘vgs‘ command.

# vgs
# lvs
Check LVM Disk Space

Check LVM Disk Space

You see, there is 8GB of free space left in above vgs output. So, let’s create a snapshot for one of my volume named tecmint_datas. For demonstration purpose, I am going to create only 1GB snapshot volume using following commands.

# lvcreate -L 1GB -s -n tecmint_datas_snap /dev/vg_tecmint_extra/tecmint_datas        

OR

# lvcreate --size 1G --snapshot --name tecmint_datas_snap /dev/vg_tecmint_extra/tecmint_datas

Both the above commands does the same thing:

  1. -s – Creates Snapshot
  2. -n – Name for snapshot
Create LVM Snapshot

Create LVM Snapshot

Here, is the explanation of each point highlighted above.

  1. Size of snapshot Iam creating here.
  2. Creates snapshot.
  3. Creates name for the snapshot.
  4. New snapshots name.
  5. Volume which we are going to create a snapshot.

If you want to remove a snapshot, you can use ‘lvremove‘ command.

# lvremove /dev/vg_tecmint_extra/tecmint_datas_snap

Remove LVM Snapshot

Remove LVM Snapshot

Now, list the newly created snapshot using following command.

# lvs
Verify LVM Snapshot

Verify LVM Snapshot

You see above, a snapshot was created successfully. I have marked with an arrow where snapshots origin from where its created, Its tecmint_datas. Yes, because we have created a snapshot for tecmint_datas l-volume.

Check LVM Snapshot Space

Check LVM Snapshot Space

Let’s add some new files into tecmint_datas. Now volume has some data’s around 650MB and our snapshot size is 1GB. So there is enough space to backup our changes in snap volume. Here we can see, what is the status of our snapshot using below command.

# lvs
Check Snapshot Status

Check Snapshot Status

You see, 51% of snapshot volume was used now, no issue for more modification in your files. For more detailed information use command.

# lvdisplay vg_tecmint_extra/tecmint_data_snap
View Snapshot Information

View Snapshot Information

Again, here is the clear explanation of each point highlighted in the above picture.

  1. Name of Snapshot Logical Volume.
  2. Volume group name currently under use.
  3. Snapshot volume in read and write mode, we can even mount the volume and use it.
  4. Time when the snapshot was created. This is very important because snapshot will look for every changes after this time.
  5. This snapshot belongs to tecmint_datas logical volume.
  6. Logical volume is online and available to use.
  7. Size of Source volume which we took snapshot.
  8. Cow-table size = copy on Write, that means whatever changes was made to the tecmint_data volume will be written to this snapshot.
  9. Currently snapshot size used, our tecmint_datas was 10G but our snapshot size was 1GB that means our file is around 650 MB. So what its now in 51% if the file grow to 2GB size in tecmint_datas size will increase more than snapshot allocated size, sure we will be in trouble with snapshot. That means we need to extend the size of logical volume (snapshot volume).
  10. Gives the size of chunk for snapshot.

Now, let’s copy more than 1GB of files in tecmint_datas, let’s see what will happen. If you do, you will get error message saying ‘Input/output error‘, it means out of space in snapshot.

Add Files to Snapshot

Add Files to Snapshot

If the logical volume become full it will get dropped automatically and we can’t use it any more, even if we extend the size of snapshot volume. It is the best idea to have the same size of Source while creating a snapshot, tecmint_datas size was 10G, if I create a snapshot size of 10GB it will never over flow like above because it has enough space to take snap of your volume.

Step 2: Extend Snapshot in LVM

If we need to extend the snapshot size before overflow we can do it using.

# lvextend -L +1G /dev/vg_tecmint_extra/tecmint_data_snap

Now there was totally 2GB size for snapshot.

Extend LVM Snapshot

Extend LVM Snapshot

Next, verify the new size and COW table using following command.

# lvdisplay /dev/vg_tecmint_extra/tecmint_data_snap

To know the size of snap volume and usage %.

# lvs
Check Size of Snapshot

Check Size of Snapshot

But if, you have snapshot volume with the same size of Source volume we don’t need to worry about these issues.

Step 3: Restoring Snapshot or Merging

To restore the snapshot, we need to un-mount the file system first.

# unmount /mnt/tecmint_datas/
Un-mount File System

Un-mount File System

Just check for the mount point whether its unmounted or not.

# df -h
Check File System Mount Points

Check File System Mount Points

Here our mount has been unmounted, so we can continue to restore the snapshot. To restore the snap using command lvconvert.

# lvconvert --merge /dev/vg_tecmint_extra/tecmint_data_snap
Restore LVM Snapshot

Restore LVM Snapshot

After the merge is completed, snapshot volume will be removed automatically. Now we can see the space of our partition using df command.

# df -Th
Check Size of Snapshot

Check Size of Snapshot

After the snapshot volume removed automatically. You can see the size of logical volume.

# lvs
Check Size of Logical Volume

Check Size of Logical Volume

Important: To Extend the Snapshots automatically, we can do it using some modification in conf file. For manual we can extend using lvextend.

Open the lvm configuration file using your choice of editor.

# vim /etc/lvm/lvm.conf

Search for word autoextend. By Default the value will be similar to below.

LVM Configuration

LVM Configuration

Change the 100 to 75 here, if so auto extend threshold is 75 and auto extend percent is 20, it will expand the size more by 20 Percent

If the snapshot volume reach 75% it will automatically expand the size of snap volume by 20% more. Thus,we can expand automatically. Save and exit the file using wq!.

This will save snapshot from overflow drop. This will also help you to save more time. LVM is the only Partition method in which we can expand more and have many features as thin Provisioning, Striping, Virtual volume and more Using thin-pool, let us see them in the next topic.

If You Appreciate What We Do Here On TecMint, You Should Consider:

TecMint is the fastest growing and most trusted community site for any kind of Linux Articles, Guides and Books on the web. Millions of people visit TecMint! to search or browse the thousands of published articles available FREELY to all.

If you like what you are reading, please consider buying us a coffee ( or 2 ) as a token of appreciation.

Support Us

We are thankful for your never ending support.

Babin Lonston

I'm Working as a System Administrator for last 10 year's with 4 years experience with Linux Distributions, fall in love with text based operating systems.

Your name can also be listed here. Got a tip? Submit it here to become an TecMint author.

RedHat RHCE and RHCSA Certification Book
Linux Foundation LFCS and LFCE Certification Preparation Guide

You may also like...

22 Responses

  1. vinci says:

    “It is the best idea to have the same size of Source while creating a snapshot, tecmint_datas size was 10G” Generally I think this is really bad advice, especially if you want to do a backup.

    If you have, let’s say, 1TB of data that you want a snapshot of, and of which let’s say a maximum of 1% (10GB) is going to change during the copying of a snapshot, then you don’t want that snapshot to be no more than 10GB.

    Saying it is ‘the best idea’ to allocate the whole size of the source is really not clever. That would mean you’d need to have yet another Terabyte in the volume group just for doing a backup, let alone space you take up with the backup itself.

  2. Alex says:

    After the step to create the snapshot via “lvcreate”, the next step says to “Let’s add some new files into tecmint_datas”, but you cannot copy anything to it because it has not been mounted yet.

    I think you’re missing the step of mounting the snapshot.

    • pepa65 says:

      The snapshot is not tecmint_datas (but tecmint_datas_snap), and indeed the snapshot is not mounted at this point. But the volume that the snapshot was taken off (tecmint_datas) was and is still mounted, and the point is to move some files to the ‘original‘, which doesn’t affect the snapshot. But if the amount of changes is larger than the storage space allows, the snapshot is (silently) dropped.

  3. marc DECHICO says:

    Of course I have limited experience and knowledge in system admin and LVM, but this is the kind of additional info that I would like to find about LVM snapshot concept in your page:

    typical use

    create the snapshot in few seconds without having to umount the volume that is snapshoted and backup immediately the system without having to stop access to system-users during the backup.
    possible use (? because I have no experience) :

    create the snapshot it in order to have the possibility to return back very quickly by mounting read-write the snapshot itself. I mean: to avoid merging.

    create some kind of version control using snapshot like branches (of course several snapshots used simultaneously will increase dramatically IO traffic.

    hope it will help.

Leave a Reply to Alex Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.