Chapter 10. Snapshot of logical volumes

Procedure

1. Create a logical volume named origin from the volume group vg001:

   # lvcreate -L 1G -n origin vg001
   Logical volume "origin" created.

2. Create a snapshot logical volume named snap of /dev/vg001/origin that is 100 MB in size:

   # lvcreate --size 100M --name snap --snapshot /dev/vg001/origin
     Logical volume "snap" created.

   You can also use the -L argument instead of using --size, -n instead of using --name, and -s instead of using --snapshot to create a snapshot.

   If the original logical volume contains a file system, you can mount the snapshot logical volume on an arbitrary directory in to access the contents of the file system to run a backup while the original file system continues to get updated.

3. Display the origin volume and the current percentage of the snapshot volume being used:

   # lvs -a -o +devices
     LV      VG    Attr       LSize  Pool Origin Data% Meta% Move Log Cpy%Sync Convert Devices
    origin vg001  owi-a-s---  1.00g                                                  /dev/sde1(0)
     snap vg001  swi-a-s--- 100.00m     origin 0.00                                 /dev/sde1(256)

   You can also display the status of logical volume /dev/vg001/origin with all the snapshot logical volumes and their status, such as active or inactive by using the lvdisplay /dev/vg001/origin command.

   Warning

   Space in the snapshot LV is consumed after the origin LV is written to. The lvs command reports the current snapshot space usage in the Data% data_percent field value. If the snapshot space reaches 100%, the snapshot becomes invalid and unusable.

   An invalid snapshot is reported with I in the fifth position of the Attr column, or the lv_snapshot_invalid reporting field in lvs. You can remove the invalid snapshot by using the lvremove command.

4. Optional: Extend the snapshot before its space becomes 100% full and becomes invalid by using any one of the following options:

   • Configure LVM to automatically extend the snapshot by using the following parameters in the /etc/lvm.conf file:

     snapshot_autoextend_threshold

     Extends the snapshot after its usage exceeds the value set for this parameter. By default, it is set to 100, which disables automatic extension. The minimum value of this parameter is 50.

     snapshot_autoextend_percent

     Adds an additional space to the snapshot, which is the percent of its current size. By default, it is set to 20.

     In the following example, after setting the following parameters, the created 1G snapshot extends to 1.2G when its usage exceeds 700M:

     Example 10.1. Automatically extend the snapshot

     # vi /etc/lvm.conf
     snapshot_autoextend_threshold = 70
     snapshot_autoextend_percent = 20

     Note

     This feature requires unallocated space in the volume group. An automatic extension of a snapshot does not increase the size of a snapshot volume beyond the maximum calculated size that is necessary for the snapshot. Once a snapshot has grown large enough to cover the origin, it is no longer monitored for automatic extension.

   • Extend this snapshot manually by using the lvextend command:

     # lvextend -L+100M /dev/vg001/snap

Procedure

1. Merge the snapshot volume. The following command merges snapshot volume vg001/snap into its origin:

   # lvconvert --merge vg001/snap
   Merging of volume vg001/snap started.
     vg001/origin: Merged: 100.00%

2. View the origin volume:

   # lvs -a -o +devices
     LV      VG    Attr       LSize  Pool Origin Data% Meta% Move Log Cpy%Sync Convert Devices
     origin vg001  owi-a-s---  1.00g                                                  /dev/sde1(0)

Procedure

1. Create a playbook file, for example ~/playbook.yml, with the following content:

   ---
   - name: Run the snapshot system role
     hosts: managed-node-01.example.com
     vars:
       snapshot_lvm_set:
         name: snapset1
         volumes:
           - name: data1 snapshot
             vg: data_vg
             lv: data1
             percent_space_required: 25
             mountpoint: /data1_snapshot
             options: nouuid
             mountpoint_create: true
           - name: data2 snapshot
             vg: data_vg
             lv: data2
             percent_space_required: 25
             mountpoint: /data2_snapshot
             options: nouuid
             mountpoint_create: true
     tasks:
       - name: Create a snapshot set
         ansible.builtin.include_role:
           name: rhel-system-roles.snapshot
         vars:
           snapshot_lvm_action: snapshot
       - name: Verify the set of snapshots for the LVs
         ansible.builtin.include_role:
           name: rhel-system-roles.snapshot
         vars:
           snapshot_lvm_action: check
           snapshot_lvm_verify_only: true
       - name: Mount the snapshot set
         ansible.builtin.include_role:
           name: rhel-system-roles.snapshot
         vars:
           snapshot_lvm_action: mount

   Here, the snapshot_lvm_set parameter describes specific logical volumes (LV) from the same volume group (VG). You can also specify LVs from different VGs while setting this parameter.

2. Validate the playbook syntax:

   $ ansible-playbook --syntax-check ~/playbook.yml

   Note that this command only validates the syntax and does not protect against a wrong but valid configuration.

3. Run the playbook:

   $ ansible-playbook ~/playbook.yml

Procedure

1. Create a playbook file, for example ~/playbook.yml, with the following content:

   • Unmount a specific LVM snapshot:

     ---
     - name: Unmount the snapshot specified by the snapset
       hosts: managed-node-01.example.com
       vars:
         snapshot_lvm_snapset_name: snapset1
         snapshot_lvm_action: umount
         snapshot_lvm_vg: data_vg
         snapshot_lvm_lv: data2
         snapshot_lvm_mountpoint: /data2_snapshot
       roles:
         - rhel-system-roles.snapshot

     Here, the snapshot_lvm_lv parameter describes a specific logical volume (LV) and the snapshot_lvm_vg parameter describes a specific volume group (VG).

   • Unmount a set of LVM snapshots:

     ---
     - name: Unmount a set of snapshots
       hosts: managed-node-01.example.com
       vars:
         snapshot_lvm_action: umount
         snapshot_lvm_set:
           name: snapset1
           volumes:
             - name: data1 snapshot
               vg: data_vg
               lv: data1
               mountpoint: /data1_snapshot
             - name: data2 snapshot
               vg: data_vg
               lv: data2
               mountpoint: /data2_snapshot
       roles:
         - rhel-system-roles.snapshot

     Here, the snapshot_lvm_set parameter describes specific LVs from the same VG. You can also specify LVs from different VGs while setting this parameter.

2. Validate the playbook syntax:

   $ ansible-playbook --syntax-check ~/playbook.yml

   Note that this command only validates the syntax and does not protect against a wrong but valid configuration.

3. Run the playbook:

   $ ansible-playbook ~/playbook.yml

Procedure

1. Create a playbook file, for example ~/playbook.yml, with the following content:

   • Extend all LVM snapshots by specifying the value for the percent_space_required parameter:

     ---
     - name: Extend all snapshots
       hosts: managed-node-01.example.com
       vars:
         snapshot_lvm_action: extend
         snapshot_lvm_set:
           name: snapset1
           volumes:
             - name: data1 snapshot
               vg: data_vg
               lv: data1
               percent_space_required: 40
             - name: data2 snapshot
               vg: data_vg
               lv: data2
               percent_space_required: 40
       roles:
         - rhel-system-roles.snapshot

     Here, the snapshot_lvm_set parameter describes specific LVs from the same VG. You can also specify LVs from different VGs while setting this parameter.

   • Extend a LVM snapshot set by setting percent_space_required to different value for each VG and LV pair in a set:

     ---
     - name: Extend the snapshot
       hosts: managed-node-01.example.com
       vars:
         snapshot_extend_set:
           name: snapset1
           volumes:
             - name: data1 snapshot
               vg: data_vg
               lv: data1
               percent_space_required: 30
             - name: data2 snapshot
               vg: data_vg
               lv: data2
               percent_space_required: 40
       tasks:
         - name: Extend data1 to 30% and data2 to 40%
           vars:
             snapshot_lvm_set: "{{ snapshot_extend_set }}"
             snapshot_lvm_action: extend
           ansible.builtin.include_role:
             name: rhel-system-roles.snapshot

2. Validate the playbook syntax:

   $ ansible-playbook --syntax-check ~/playbook.yml

   Note that this command only validates the syntax and does not protect against a wrong but valid configuration.

3. Run the playbook:

   $ ansible-playbook ~/playbook.yml

Procedure

1. Create a playbook file, for example ~/playbook.yml, with the following content:

   • Revert a specific LVM snapshot to its original volume:

     ---
     - name: Revert a snapshot to its original volume
       hosts: managed-node-01.example.com
       vars:
         snapshot_lvm_snapset_name: snapset1
         snapshot_lvm_action: revert
         snapshot_lvm_vg: data_vg
         snapshot_lvm_lv: data2
       roles:
         - rhel-system-roles.snapshot

     Here, the snapshot_lvm_lv parameter describes a specific logical volume (LV) and the snapshot_lvm_vg parameter describes a specific volume group (VG).

   • Revert a set of LVM snapshots to its original volume:

     ---
     - name: Revert a set of snapshot
       hosts: managed-node-01.example.com
       vars:
         snapshot_lvm_action: revert
         snapshot_lvm_set:
           name: snapset1
           volumes:
             - name: data1 snapshot
               vg: data_vg
               lv: data1
             - name: data2 snapshot
               vg: data_vg
               lv: data2
       roles:
         - rhel-system-roles.snapshot

     Here, the snapshot_lvm_set parameter describes specific LVs from the same VG. You can also specify LVs from different VGs while setting this parameter.

     Note

     The revert operation might take some time to complete.

2. Validate the playbook syntax:

   $ ansible-playbook --syntax-check ~/playbook.yml

   Note that this command only validates the syntax and does not protect against a wrong but valid configuration.

3. Run the playbook:

   $ ansible-playbook ~/playbook.yml

4. Reboot the host, or deactivate and reactivate the logical volume using the following steps:

   $ umount /data1; umount /data2
   
   $ lvchange -an data_vg/data1 data_vg/data2
   
   $ lvchange -ay data_vg/data1 data_vg/data2
   
   $ mount /data1; mount /data2

Procedure

1. Create a playbook file, for example ~/playbook.yml, with the following content:

   • Remove a specific LVM snapshot:

     ---
     - name: Remove a snapshot
       hosts: managed-node-01.example.com
       vars:
         snapshot_lvm_snapset_name: snapset1
         snapshot_lvm_action: remove
         snapshot_lvm_vg: data_vg
       roles:
         - rhel-system-roles.snapshot

     Here, the snapshot_lvm_vg parameter describes a specific logical volume (LV) from the volume group (VG).

   • Remove a set of LVM snapshots:

     ---
     - name: Remove a set of snapshots
       hosts: managed-node-01.example.com
       vars:
         snapshot_lvm_action: remove
         snapshot_lvm_set:
           name: snapset1
           volumes:
             - name: data1 snapshot
               vg: data_vg
               lv: data1
             - name: data2 snapshot
               vg: data_vg
               lv: data2
       roles:
         - rhel-system-roles.snapshot

     Here, the snapshot_lvm_set parameter describes specific LVs from the same VG. You can also specify LVs from different VGs while setting this parameter.

2. Validate the playbook syntax:

   $ ansible-playbook --syntax-check ~/playbook.yml

   Note that this command only validates the syntax and does not protect against a wrong but valid configuration.

3. Run the playbook:

   $ ansible-playbook ~/playbook.yml