LVM
source: https://unixutils.com/lvm-cheat-sheet-quick-reference/
CREATE PV,VG,LV
pvcreate /dev/sda1 /dev/sdb1 vgcreate vg_apps /dev/sda1 /dev/sdb1 #create LV with all available space from vg_apps lvcreate -n lv_apps -l 100%FREE vg_apps #create LV with specific size lvcreate -L 108G -n lv_apps vg_apps mkfs.ext4 /dev/vg_apps/lv_apps
Other FS-TYPES: ext2,ext3,ext4,xfs,btrfs,cramfs,minix
Display PV,VG,LV details
pvdisplay pvs vgdisplay vgs lvdisplay lvs
EXTEND VG,LV
#Extends VG with a new Disk (after pvcreate command has been run the disk eg: pvcreate /dev/sdj1) vgextend vg_apps /dev/sdj1 #Extends LV by 100GB lvextend -L +100G /dev/mapper/vg_apps-lv_apps #Extends LV with all available space in vg_apps lvextend -l +100%FREE /dev/mapper/vg_apps-lv_apps
CREATE, RESTORE LVM SNAPSHOT
#create a snapshot of LV: lv_apps with size 1G lvcreate -L 1GB -s -n my_apps_backup /dev/vg_apps/lv_apps /dev/vg_apps/lv_apps_backup #restore a snapsot lvconvert --merge /dev/vg_apps/lv_apps_backup #remove a snapsot lvremove /dev/vg_apps/lv_apps_backup
NOTE: how to choose snapshot size ?. Snapshots are used to store changes in an LVM from the point it was created. So, let's say that you have a 30GB volume and you take a snapshot. Initially, no space is required. But after a week you see that the snapshot is using 4GB of storage. This means that Size or amount of changes since the point of snapshot creation is is 4G.
To restore the snapshot, we need to un-mount the file system first. And, after the merge is completed, snapshot volume will be removed automatically.
DEALING WITH FAILED VOLUMES AND RECOVERY
Dealing with “Couldn’t find device with uuid ‘FmGRh3-zhok-iVI8-7qTD-S5BI-MAEN-NYM5Sk'” errors.
This can be a case of an actual missing disk or If the volume group metadata area of a physical volume is accidentally overwritten or otherwise destroyed, you will get an error message indicating that the metadata area is incorrect, or that the system was unable to find a physical volume with a particular UUID. You may be able to recover the data the physical volume by writing a new metadata area on the physical volume specifying the same UUID as the lost metadata.
LVM related files and directories: /etc/lvm/ - default lvm directory location /etc/lvm/backup/ - Contains last backup of metadata for VG /etc/lvm/archive/ - Contains backup of metadata for VG before something changed /etc/lvm/lvm.conf - main lvm configuration file
Check for failed volumes
vgs -a -o +devices -P lvs -a -o +devices -P
Couldn't find device with uuid 'DDGRh3-zhok-iVI8-7qTD-S5BI-GAEN-NYM5Sk'.
Couldn't find all physical volumes for volume group vg_apps. Couldn't find device with uuid 'DDGRh3-zhok-iVI8-7qTD-S5BI-GAEN-NYM5Sk'. Couldn't find all physical volumes for volume group vg_apps. ...
Restore Metadata from backup
pvcreate --uuid "DDGRh3-zhok-iVI8-7qTD-S5BI-GAEN-NYM5Sk" --restorefile /etc/lvm/archive/vg_apps_00050.vg /dev/sdh1 vgcfgrestore vg_apps lvchange -ay /dev/vg_apps/lv_apps
Mirroring
Show mirrors:
# lvs -a -o +devices
Create mirroring on new pv:
# lvconvert -m2 /dev/datavg/libdata /dev/sda
- m0 = single, no mirroring
- m1 = data + one mirror
- m2 = data + 2 mirrors
Remove missing volumes in case of failure:
# vgreduce --removemissing volgroup_mirror --force