How to fix: At least xMB more space needed on the /boot filesystem

Running sudo dnf upgrade in Fedora CentOS or RHEL will sometimes result in the following error: At least xxMB more space is needed on the /boot filesystem. Let’s fix this!

List all installed kernel packages with:

yum list kernel

On Fedora 22+, CentOS Stream and RHEL 8+, use:

dnf list kernel

The kernel in use will be underlined and cannot be removed:
yum list kernel
Now to remove unused kernels, install the yum-utils (or dnf-utils) package and use the package-cleanup util:

yum install yum-utils
package-cleanup --oldkernels --count=2

To make this permanent, edit /etc/yum.conf and add the following line:

installonly_limit=2

On RHEL8/CentOS 8/stream, use:

dnf remove --oldinstallonly --setopt installonly_limit=2 kernel

– – – – – – – – – – – – – – – – – – – – — – – – – – – – – – – – – – – – – – – – — – – – – – – – – – – – – – – – – – – – –

The original article from 2017:

While updating a client’s CentOS 6 server using “yum update” I ran into the following error:

Transaction Check Error: installing package kernel-2.6.32-358.2.1.el6.x86_64 needs 11MB on the /boot filesystem
Error Summary
Disk Requirements: At least 11MB more space needed on the /boot filesystem.

To check the free space on /boot I ran the following command: df -h

Which gave the following output:

Filesystem Size Used Avail Use% Mounted on
/dev/vda3 136G 60G 70G 46% /
tmpfs 5.3G 0 5.3G 0% /dev/shm
/dev/vda1 99M 77M 17M 82% /boot
/usr/tmpDSK 1.3G 35M 1.2G 3% /tmp

As you can see the boot partition is 82% full.

To list installed kernels use this command:

rpm -qa | grep kernel

The result:

kernel-2.6.32-220.7.1.el6.x86_64
kernel-headers-2.6.32-358.2.1.el6.x86_64
dracut-kernel-004-303.el6.noarch
kernel-2.6.32-279.19.1.el6.x86_64
kernel-firmware-2.6.32-358.2.1.el6.noarch
kernel-2.6.32-279.9.1.el6.x86_64

So to remove the oldest kernel, I ran:

rpm -e kernel-2.6.32-220.7.1.el6.x86_64

Then: df -h

Output:

Filesystem Size Used Avail Use% Mounted on
/dev/vda3 136G 60G 70G 46% /
tmpfs 5.3G 0 5.3G 0% /dev/shm
/dev/vda1 99M 52M 42M 56% /boot
/usr/tmpDSK 1.3G 35M 1.2G 3% /tmp

As you can see that freed up more than the 11MB needed. So I retried yum -y update kernel and this time it works. Finally, I deleted the old ones again:

rpm -e kernel-2.6.32-279.9.1.el6.x86_64 kernel-2.6.32-279.19.1.el6.x86_64

The final result, 70M free!.

/dev/vda1 99M 25M 70M 26% /boot

Last updated: Aug 13th, 2023

Tags: ,

Stop losing users to slow load times.

Blazing-fast, custom-configured NVMe Linux hosting.

Don't guess if your server is fast enough. Know it. Get a free performance audit of your current setup.

This blog is hosted by StackLinux, our parent company. 99.99% uptime.

Start Your Free Server Audit

You write the code. We'll manage the server.

Fully managed Linux hosting built for absolute speed.

Ditch the server maintenance headaches. Our experts custom-configure high-performance NVMe Linux servers specifically for your needs. Try us out. Your first month is entirely on us.

This blog is hosted by StackLinux, our parent company. 99.99% uptime.

Claim Your 30 Free Days

30 days of NVMe Linux hosting. $0 down.

Risk-free migration and a money-back guarantee.

We are so confident in our high-performance NVMe infrastructure that we'll audit your current server for free, migrate you without downtime, and give you 30 days to see the speed difference yourself.

This blog is hosted by StackLinux, our parent company. 99.99% uptime.

Deploy Free for 30 Days
Top ↑