- Ensure PKI intermediates have been created
- Make sure all services are properly deployed to staging-codfw (no surprises for deployers)
- Switch the active staging cluster to staging-codfw (https://wikitech.wikimedia.org/wiki/Kubernetes/Administration#Switch_the_active_staging_cluster_%28eqiad%3C-%3Ecodfw%29)
- Downtime: etcd, master, nodes
- Reimage etcd nodes with bullseye
- Merge hiera changes for 1.23
- Reimage master
- Reimage nodes
- Verify basic k8s stuff working (nodes joining the cluster)
- Marge deployment-charts changes for 1.23
- Deploy admin_ng & istio
- Deploy services
- Lift downtimes
- Switch the active staging cluster to staging-eqiad (https://wikitech.wikimedia.org/wiki/Kubernetes/Administration#Switch_the_active_staging_cluster_%28eqiad%3C-%3Ecodfw%29)
Detailed steps and commands can be found in T326340: Update staging-codfw to k8s 1.23