https://github.com/cgwalters/openshift-machine-bootimage-updater

https://github.com/cgwalters/openshift-machine-bootimage-updater

Last synced: 1 day ago
JSON representation

• Host: GitHub
• URL: https://github.com/cgwalters/openshift-machine-bootimage-updater
• Owner: cgwalters
• License: apache-2.0
• Created: 2021-04-28T22:04:31.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2021-04-28T22:05:08.000Z (over 3 years ago)
• Last Synced: 2024-11-16T21:41:58.268Z (about 1 month ago)
• Language: Go
• Size: 3.37 MB
• Stars: 1
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# PoC code to update bootimages in OpenShift 4

See https://github.com/openshift/enhancements/pull/201

## Status

Tested and working in GCP from a cluster installed as 4.5 and upgraded in place to 4.6.

Other platforms:

 - AWS: Code written but not tested yet

## Usage

Your cluster must be using an OpenShift release channel that ends in `-4.6`; for example, `stable-4.6`.

First, a good thing to do would be to create a backup of your machinesets:

```

$ oc -n openshift-machine-api get -o yaml machineset > machinesets.yaml

```

Let's create an admin service account for this, then run it as a pod:

```

$ oc project kube-system

$ oc create sa bootimage-updater

$ oc adm policy add-cluster-role-to-user cluster-admin -z bootimage-updater

$ oc run --restart=Never --serviceaccount=bootimage-updater --image=registry.svc.ci.openshift.org/cgwalters/openshift-update-bootimages update-bootimages 

```

Look at the logs:

```

$ oc logs pod/update-bootimages

```

If that succeeded, in order to test that this fully works, try [scaling up one of your machinesets](https://docs.openshift.com/container-platform/4.6/machine_management/manually-scaling-machineset.html).

## How it works

We first translate e.g. `worker-user-data` to Ignition spec 3x, creating a new `worker-user-data-ignv3`.

Then we loop over all the machinesets, and update them to use the new user-data secret plus the updated bootimage.

## Reverting/debugging

See [this FAQ entry](https://github.com/openshift/machine-api-operator/blob/master/FAQ.md#i-created-a-machine-but-it-never-joined-the-cluster) if the scaled up node doesn't join.

If you want to revert; at this time, we don't automatically back up the machinesets; hopefully you did that above.