Home / News / A roadmap of the OpenShift boot images update

A roadmap of the OpenShift boot images update

Red Hat is updating boot images across all our platforms. We will perform the updates on a platform-by-platform basis. We aim to initially release it as an opt-in feature. Once all the kinks have been worked out for a particular platform, we will turn on boot image updates by default for that platform. A cluster administrator could still opt out of boot image updates. For user-managed clusters and those that do not use machine resources like MachineSets, we will provide documentation to perform boot image updates. 

Eventually, cluster admins concerned about scaling will be responsible for keeping their boot images updated within a certain value to preserve scaling and upgrade ability.

Who will be impacted and when?

All customers will be impacted by this update, but IPI-based customers are most likely to see the effects first. As stated, this feature is critical for clusters that scale frequently and have been around for a while.

The following chart is a tentative timeline for when these changes will go into place, subject to change.

Platform

Tech Preview (opt-in)

GA (opt-in)

Default-On (opt-out)

GCP

4.16

4.17

4.19

AWS

4.17

4.18

4.19

vSphere

4.20

4.21

4.22

Azure

4.21

4.22

4.23

OpenStack

4.21

4.22

4.23

Nutanix

4.21

4.22

4.23

IBM Cloud

4.22

4.23

4.24

non-managed

*

*

4.23*

*Any platform that does not use MachineSets will need to update boot images manually.

Why now?

We are implementing this change because it is high time! Currently, boot image references are pinned in a Machine Resource (typically a MachineSet) by the Red Hat OpenShift installer during cluster bring-up. These boot image references are not updated on an upgrade. As a result, any new nodes scaled from this machine resource will start from this boot image reference and pivot to the Red Hat Enterprise Linux CoreOS/Red Hat Enterprise Linux image referenced by the release image.

For example, a cluster installed on 4.6 and upgraded to 4.16 will have new nodes scaled up on the 4.6 OS image and then attempt to update to the 4.16 OS image. As you can imagine, as this skew widens, this could cause issues during the final OS update to the latest OS image due to missing or incompatible packages.

The post A roadmap of the OpenShift boot images update appeared first on Red Hat Developer.

Tagged: