How to Uninstall CloudLinux (Revert to CentOS / AlmaLinux)
This guide reverts CloudLinux back to its underlying RHEL-family base — typically CentOS 7 or AlmaLinux 8 / 9 depending on which CloudLinux version was installed. The same procedure applies to all variants (CloudLinux, CloudLinux V2 / V8 / V9).
Note: Uninstalling does not cancel your billing. If you also want to cancel the license, please reply to your support ticket with the server IP — billing is handled separately.
Prerequisites
- Root SSH access to your server
- Ability to reboot the server twice — required to switch the running kernel
- A maintenance window of 30 to 60 minutes — sites will be offline during reboots
- Working backups of , your sites, MySQL, and any LVE-tuned configsTerminal
/etc
Step 1 — Backup CloudLinux-specific configs (optional but recommended)
If you have customised LVE limits, MySQL Governor rules, or PHP Selector settings, save a copy:
cp -a /etc/container /root/cloudlinux-backup-$(date +%F)/
cp -a /etc/cl.selector /root/cloudlinux-backup-$(date +%F)/ 2>/dev/nullYou won't be able to use these once CloudLinux is removed, but they're useful if you ever migrate back.
Step 2 — Run the CloudLinux revert command
This is the official vendor command and is the only supported uninstall path:
cldeploy -c -yWhat it does:
- Removes CloudLinux-specific kernels and packages
- Restores the CentOS or AlmaLinux base packages
- Updates
yumrepositories to point at the upstream repo - Schedules a kernel switch on the next boot
The script takes 5 to 15 minutes. Do not interrupt it.
Step 3 — First reboot (kernel switch)
rebootAfter it comes back up, the running kernel will be CentOS / AlmaLinux. Verify:
uname -r
# Expect something like 4.18.0-553.el8.x86_64 (AlmaLinux)
# or 3.10.0-1160.el7.x86_64 (CentOS 7)
# NOT "lve" in the kernel nameIf you still see .lve. in the kernel name, the GRUB default did not switch. Run:
grubby --set-default-index=0
rebootStep 4 — Second reboot (cleanup of LVE userspace) — only if needed
In some cases CloudLinux's userspace components (cagefs, lvectl, lve_setup) leave service stubs that fail at boot. If
systemctl --failedsystemctl disable cagefs
systemctl disable lvestats
yum remove cagefs lve-utils lvemanager mysql-governor -y
rebootStep 5 — Verify
# OS should report CentOS or AlmaLinux, NOT CloudLinux
cat /etc/os-release | grep ^NAME
# No CloudLinux packages remaining
rpm -qa | grep -iE 'cloudlinux|lve|cagefs|lvemanager|mysql-governor'
# Output should be emptyIf both checks pass, CloudLinux is fully removed.
Troubleshooting
cldeploy: command not found
CloudLinux userspace tools were partially removed. Reinstall them temporarily, then revert:
yum install cloudlinux-release cldeploy -y
cldeploy -c -y
rebootSites are down after the reboot
Restart the web server and database:
systemctl restart httpd mysqld # cPanel / standalone
systemctl restart litespeed mysqld # if LiteSpeed is still installedIf WHM accounts are missing, run
/scripts/upcp --forceMySQL won't start (mysql-governor leftover)
/usr/share/lve/dbgovernor/mysqlgovernor.py --uninstall
systemctl restart mysqld