Your submission was sent successfully! Close

Thank you for signing up for our newsletter!
In these regular emails you will find the latest updates from Canonical and upcoming events where you can meet our team.Close

Thank you for contacting our team. We will be in touch shortly.Close

  1. Blog
  2. Article

Francis Ginther
on 16 April 2020


The linux-aws 4.15 based kernel, which is the default kernel in the Ubuntu 18.04 LTS AMIs, is moving to a rolling kernel model.

Why is this changing?

The Ubuntu rolling kernel model provides the latest upstream bug fixes and performance improvements around task scheduling, I/O scheduling, networking, hypervisor guests and containers to our users. Canonical has been following this model in other cloud environments for some time now, and have found it to be an excellent way to deliver these benefits while continuing to provide LTS level stability.

What is the rolling kernel model?

A rolling kernel model transitions the default linux-aws kernel from one base version to the next as part of its regular patching cycle. That new kernel is the kernel of the latest interim Ubuntu release.  Applying this model directly to 18.04 today, the linux-aws kernel is a 4.15 based kernel and when we roll, it will become a 5.3 based kernel which was part of our 19.10 interim release.  

Today, that 5.3 kernel is currently available for preview as the linux-aws-edge kernel, which we encourage all users to run with their workloads in non-production deployments.  It is important to keep in mind that both the -edge kernels and the rolling release kernels are fully baked prior to being made available for our customers to use and meet exactly the same quality and durability standards all our kernels must meet for release. 

When the linux-aws kernel rolls forward, a user would see this change in 1 of 2 ways, 1) launching the latest AMI would have the newer kernel, and 2) users applying packaging updates, or via automatic security updates, will also see the newer kernel.

How do I prepare for the ‘roll’?

To install the 5.3 linux-aws-edge kernel, which is currently available for customer testing today, please follow these short instructions using a terminal window:

Note that the instance is running the standard linux-aws kernel (v4.15.0):

ubuntu@ip-xxx-xxx-xxx-xxx$ uname -r
4.15.0-xxxx-aws

Install the linux-aws-edge kernel and reboot the instance:

ubuntu@ip-xxx-xxx-xxx-xxx$ sudo apt update
ubuntu@ip-xxx-xxx-xxx-xxx$ sudo apt install -y linux-aws-edge

Note: If prompted about a new version of /boot/grub/menu.lst, select the default choice: “keep the local version currently installed”.

ubuntu@ip-xxx-xxx-xxx-xxx$ sudo reboot

Confirm that the instance is now running the linux-aws-edge kernel (v5.3.0):

ubuntu@ip-xxx-xxx-xxx-xxx$ uname -r
5.3.0-xxxx-aws

What if I don’t want my kernel to roll?

If you do not want to roll to a new kernel but instead stay on the 4.15 based kernel (which will continue to get full support and updates for the length of the LTS), just type the following instructions into a terminal window ssh’d into your instance:

ubuntu@ip-xxx-xxx-xxx-xxx$ sudo apt update
ubuntu@ip-xxx-xxx-xxx-xxx$ sudo apt install linux-aws-lts-18.04

How can you provide feedback?

The best way to provide feedback on Ubuntu products is to file a bug.   Filing Ubuntu Kernel Bugs

Related posts


Canonical
30 March 2023

Time to prepare for Ubuntu 18.04 LTS end of standard support on 31 May 2023 – Options for AWS users.

Cloud and server Article

As mentioned in our recent blog post, Ubuntu 18.04 LTS ‘Bionic Beaver‘ will reach the end of the standard five-year maintenance window on 31 May 2023. This means that Ubuntu 18.04 LTS servers will stop receiving updates unless you are an Ubuntu Pro subscriber. With an Ubuntu Pro subscription, your Ubuntu 18.04 LTS deployment will ...


Carlos Bravo
27 March 2024

Generative AI with Ubuntu on AWS. Part II: Text generation

AI Article

In our previous post, we discussed how to generate Images using Stable Diffusion on AWS. In this post, we will guide you through running LLMs for text generation in your own environment with a GPU-based instance in simple steps, empowering you to create your own solutions. Text generation, a trending focus in generative AI, facilitates ...


Hugo Huang
27 March 2024

Canonical at Google Next – What you need to know

Cloud and server Article

Learn how Canonical and Google Cloud are collaborating to secure and scale solutions for cloud computing at Google Next 2024. ...