Persistent block device naming
This article describes how to use persistent names for your block devices. This has been made possible by the introduction of udev and has some advantages over bus-based naming. If your machine has more than one SATA, SCSI or IDE disk controller, the order in which their corresponding device nodes are added is arbitrary. This may result in device names like
/dev/sdb switching around on each boot, culminating in an unbootable system, kernel panic, or a block device disappearing. Persistent naming solves these issues.
- Persistent naming has limits that are out-of-scope in this article. For example, while mkinitcpio may support a method, systemd may impose its own limits (e.g. FS#42884) on naming it can process during boot.
- If you are using LVM2, this article is not relevant as LVM takes care of this automatically.
Persistent naming methods
There are four different schemes for persistent naming: by-label, by-uuid, by-id and by-path. For those using disks with GUID Partition Table (GPT), two additional schemes can be used by-partlabel and by-partuuid. You can also use static device names by using Udev.
The following sections describes what the different persistent naming methods are and how they are used.
lsblk -f command can be used for viewing graphically the first persistent schemes:
$ lsblk -f
NAME FSTYPE LABEL UUID MOUNTPOINT sda ├─sda1 vfat CBB6-24F2 /boot ├─sda2 ext4 SYSTEM 0a3407de-014b-458b-b5c1-848e92a327a3 / ├─sda3 ext4 DATA b411dc99-f0a0-4c87-9e05-184977be8539 /home └─sda4 swap f9fe0b69-a280-415d-a03a-a32752370dee [SWAP]
For those using GPT, use the
blkid command instead. The latter is more convenient for scripts, but more difficult to read.
/dev/sda1: UUID="CBB6-24F2" TYPE="vfat" PARTLABEL="EFI SYSTEM PARTITION" PARTUUID="d0d0d110-0a71-4ed6-936a-304969ea36af" /dev/sda2: LABEL="SYSTEM" UUID="0a3407de-014b-458b-b5c1-848e92a327a3" TYPE="ext4" PARTLABEL="GNU/LINUX" PARTUUID="98a81274-10f7-40db-872a-03df048df366" /dev/sda3: LABEL="DATA" UUID="b411dc99-f0a0-4c87-9e05-184977be8539" TYPE="ext4" PARTLABEL="HOME" PARTUUID="7280201c-fc5d-40f2-a9b2-466611d3d49e" /dev/sda4: UUID="f9fe0b69-a280-415d-a03a-a32752370dee" TYPE="swap" PARTLABEL="SWAP" PARTUUID="039b6c1c-7553-4455-9537-1befbc9fbc5b"
Almost every filesystem type can have a label. All your partitions that have one are listed in the
/dev/disk/by-label directory. This directory is created and destroyed dynamically, depending on whether you have partitions with labels attached.
$ ls -l /dev/disk/by-label
total 0 lrwxrwxrwx 1 root root 10 May 27 23:31 DATA -> ../../sda3 lrwxrwxrwx 1 root root 10 May 27 23:31 SYSTEM -> ../../sda2
The labels of your filesystems can be changed. Following are some methods for changing labels on common filesystems:
swaplabel -L <label> /dev/XXXusing
e2label /dev/XXX <label>using
btrfs filesystem label /dev/XXX <label>using
reiserfstune -l <label> /dev/XXXusing
jfs_tune -L <label> /dev/XXXusing
xfs_admin -L <label> /dev/XXXusing
fatlabel /dev/XXX <label>using
mlabel -i /dev/XXX ::<label>using
exfatlabel /dev/XXX <label>using
ntfslabel /dev/XXX <label>using
- this filesystem does not support
/dev/disk/by-label, but #by-partlabel may be used
- Changing the filesystem label of the root partition has to be done from a "live" GNU/Linux distribution because the partition needs to be unmounted first.
- Labels have to be unambiguous to prevent any possible conflicts.
- Labels can be up to 16 characters long.
- Since the label is a property of the filesystem, it is not suitable for addressing a single RAID device persistently.
- When using encrypted containers with dm-crypt, the labels of filesystems inside of containers are not available while the container is locked/encrypted.
UUID is a mechanism to give each filesystem a unique identifier. These identifiers are generated by filesystem utilities (e.g.
mkfs.*) when the partition gets formatted and are designed so that collisions are unlikely. All GNU/Linux filesystems (including swap and LUKS headers of raw encrypted devices) support UUID. FAT and NTFS filesystems (fat and windows labels above) do not support UUID, but are still listed in
/dev/disk/by-uuid/ with a shorter UID (unique identifier):
$ ls -l /dev/disk/by-uuid/
total 0 lrwxrwxrwx 1 root root 10 May 27 23:31 0a3407de-014b-458b-b5c1-848e92a327a3 -> ../../sda2 lrwxrwxrwx 1 root root 10 May 27 23:31 b411dc99-f0a0-4c87-9e05-184977be8539 -> ../../sda3 lrwxrwxrwx 1 root root 10 May 27 23:31 CBB6-24F2 -> ../../sda1 lrwxrwxrwx 1 root root 10 May 27 23:31 f9fe0b69-a280-415d-a03a-a32752370dee -> ../../sda4
The advantage of using the UUID method is that it is much less likely that name collisions occur than with labels. Further, it is generated automatically on creation of the filesystem. It will, for example, stay unique even if the device is plugged into another system (which may perhaps have a device with the same label).
The disadvantage is that UUIDs make long code lines hard to read and break formatting in many configuration files (e.g. fstab or crypttab). Also every time a partition is resized or reformatted a new UUID is generated and configs have to get adjusted (manually).
by-id and by-path
by-id creates a unique name depending on the hardware serial number,
by-path depending on the shortest physical path (according to sysfs). Both contain strings to indicate which subsystem they belong to (i.e.
by-id), so they are linked to the hardware controlling the device. This implies different levels of persistence: the
by-path will already change when the device is plugged into a different port of the controller, the
by-id will change when the device is plugged into a port of a hardware controller subject to another subsystem.  Thus, both are not suitable to achieve persistent naming tolerant to hardware changes.
However, both provide important information to find a particular device in a large hardware infrastructure. For example, if you do not manually assign persistent labels (
by-partlabel) and keep a directory with hardware port usage,
by-path can be used to find a particular device. 
$ ls -l /dev/disk/by-id/
total 0 lrwxrwxrwx 1 root root 10 May 27 23:31 ata-VBOX_HARDDISK_VBe4474842-dc40379f -> ../../sda lrwxrwxrwx 1 root root 10 May 27 23:31 ata-VBOX_HARDDISK_VBe4474842-dc40379f-part1 -> ../../sda1 lrwxrwxrwx 1 root root 10 May 27 23:31 ata-VBOX_HARDDISK_VBe4474842-dc40379f-part2 -> ../../sda2 lrwxrwxrwx 1 root root 10 May 27 23:31 ata-VBOX_HARDDISK_VBe4474842-dc40379f-part3 -> ../../sda3 lrwxrwxrwx 1 root root 10 May 27 23:31 ata-VBOX_HARDDISK_VBe4474842-dc40379f-part4 -> ../../sda4
$ ls -l /dev/disk/by-path/
total 0 lrwxrwxrwx 1 root root 10 May 27 23:31 pci-0000:00:1f.2-ata-1 -> ../../sda lrwxrwxrwx 1 root root 10 May 27 23:31 pci-0000:00:1f.2-ata-1-part1 -> ../../sda1 lrwxrwxrwx 1 root root 10 May 27 23:31 pci-0000:00:1f.2-ata-1-part2 -> ../../sda2 lrwxrwxrwx 1 root root 10 May 27 23:31 pci-0000:00:1f.2-ata-1-part3 -> ../../sda3 lrwxrwxrwx 1 root root 10 May 27 23:31 pci-0000:00:1f.2-ata-1-part4 -> ../../sda4
Partition labels can be defined in the header of the partition entry on GPT disks.
This method is very similar to the filesystem labels, excepted that the dynamic directory is
ls -l /dev/disk/by-partlabel/
total 0 lrwxrwxrwx 1 root root 10 May 27 23:31 EFI\x20SYSTEM\x20PARTITION -> ../../sda1 lrwxrwxrwx 1 root root 10 May 27 23:31 GNU\x2fLINUX -> ../../sda2 lrwxrwxrwx 1 root root 10 May 27 23:31 HOME -> ../../sda3 lrwxrwxrwx 1 root root 10 May 27 23:31 SWAP -> ../../sda4
- GPT partition labels have also to be different to avoid conflicts. To change your partition label, you can use
gdiskor the ncurses-based version
cgdisk. Both are available from the package. See Partitioning#Partitioning tools.
- According to the specification, GPT partition labels can be up to 72 characters long.
Like GPT partition labels, GPT partition UUID are defined in the partition entry on GPT disks.
The dynamic directory is similar to other methods and, like UUID filesystems, using UUIDs is prefered over labels.
ls -l /dev/disk/by-partuuid/
total 0 lrwxrwxrwx 1 root root 10 May 27 23:31 039b6c1c-7553-4455-9537-1befbc9fbc5b -> ../../sda4 lrwxrwxrwx 1 root root 10 May 27 23:31 7280201c-fc5d-40f2-a9b2-466611d3d49e -> ../../sda3 lrwxrwxrwx 1 root root 10 May 27 23:31 98a81274-10f7-40db-872a-03df048df366 -> ../../sda2 lrwxrwxrwx 1 root root 10 May 27 23:31 d0d0d110-0a71-4ed6-936a-304969ea36af -> ../../sda1
Static device names with Udev
Using persistent naming
There are various applications that can be configured using persistent naming. Following are some examples of how to configure them.
See the main article: fstab#UUIDs
To use persistent names in the boot manager (boot loader), the following prerequisites must be met. On a standard installation following the installation guide both prerequisites are met.
- You are using a mkinitcpio initial RAM disk image
- You have udev enabled in
The location of the root filesystem is given by the parameter
root on the kernel commandline. The kernel commandline is configured from the bootloader, see Kernel parameters#Configuration. To change to persistent device naming, only change the parameters which specify block devices, e.g.
resume, while leaving other parameters as is. Various naming schemes are supported:
Non-persistent device naming using the
/dev path format, in this example
/dev/sdr1 is the root partition.
Persistent device naming using label and the
/dev path format, in this example
root_myhost is the label of the root partition.
Persistent device naming using label and the
Persistent device naming using UUID and the
UUID= format, in this example
1234-5678 is the UUID of the root partition.
Persistent device naming using the PARTUUID and the
Persistent device naming using the PARTLABEL and the