Persistent Mounts with UUID and Label
The objective that can cost every other mark on the exam: a malformed /etc/fstab stops the machine booting, and the grader reboots before marking. So this ends on verification - a deliberately broken entry is caught here by findmnt --verify and mount -a, which is exactly the check to run before rebooting.
Local Storage Guide 35 of 67 Intermediate
- OSRHEL 10.0
- Kernel6.12.0-55.9.1.el10_0
- dnf4.20.0
- Flatpak1.16.0
- TimeAbout 16 min
After editing /etc/fstab, run systemctl daemon-reload. systemd generates a mount unit per entry at boot and caches them; findmnt --verify warns explicitly when the cached copy is stale, as it does below.
| Server Name | IP Address | OS | Roles | CPU | RAM | HDD |
|---|---|---|---|---|---|---|
| RHCSA-A01 | 192.168.0.31 | RHEL 10.0 | Practice node (graded) - spare /dev/sda | 2 Core | 4 GB | 50 GB + 15 GB |
This guide includes
Use this because a malformed /etc/fstab can cost every other mark on the exam. This matters because a device name can move between boots - and verifying before you reboot is the difference between a pass and an unbootable machine.
- mounting a filesystem by hand first
- seeing why a device name is the wrong identifier to trust
- mounting by UUID, and by label
- reading the six fields of an fstab line
- verifying before you reboot, and using
nofailas the escape hatch
Before you start
- A logical volume or partition to mount -
guide 34makes one. - The session appends to
/etc/fstaband removes its own lines afterwards.
-
A filesystem, mounted by hand
-
Why not the device name
-
By UUID, and by label
-
The six fields
-
Verifying before you reboot
-
The nofail escape hatch