Changelog for Kubernetes 1.22

Versions

• Kubernetes 1.22.8
• Nginx-ingress: 1.1.1
• Certmanager: 1.6.3

Major changes

• When our ingress is installed we set it as the default ingress, meaning it will be used unless a custom ingress class is used/specified
• Clusters are now running containerd instead of docker. This should not affect your workload at all
• We reserve 5% RAM on all nodes making it easier to calculate how much is left for your workload
• All components deployed by Elastx have tolerations for NoSchedule taints by default
• Certmanager cert-manager.io/v1alpha2, cert-manager.io/v1alpha3, cert-manager.io/v1beta1, acme.cert-manager.io/v1alpha2, acme.cert-manager.io/v1alpha3 and acme.cert-manager.io/v1beta1 APIs are no longer served. All existing resources will be converted automatically to cert-manager.io/v1 and acme.cert-manager.io/v1, however you will still need to update your local manifests
• Several old APIs are no longer served. A complete list can be found in Kubernetes documentation

Changes affecting new clusters:

• All new clusters will have the cluster domain cluster.local by default
• The encrypted *-enc storage-classes (4k-enc, 8k-enc and 16k-enc) are no longer available to new clusters since they are deprecated for removal in Openstack. Do not worry, all our other storage classes (4k, 8k, 16k and future classes) are now encrypted by default. Read our full announcement here

Deprecations

Note that all deprecations will be removed in a future Kubernetes release. This does not mean you need to perform any changes right now. However, we recommend you to start migrating your applications in order to avoid issues in future releases.

APIs removed in Kubernetes 1.25

More details can be found in Kubernetes official documentation.

• Pod Security Policies will be removed in Kubernetes 1.25
• CronJob batch/v1beta1. The new API batch/v1 was implemented in Kubernetes 1.21 (this is a drop in replacement)
• EndpointSlice discovery.k8s.io/v1beta1. The new API discovery.k8s.io/v1 was implemented in Kubernetes 1.21
• Event events.k8s.io/v1beta1. The new API events.k8s.io/v1 was implemented in Kubernetes 1.19
• PodDisruptionBudget policy/v1beta1. The new API policy/v1 was implemented in Kubernetes 1.21
• RuntimeClass node.k8s.io/v1beta1. The new API node.k8s.io/v1 was implemented in Kubernetes 1.20

Other noteworthy deprecations

Kubernetes beta topology labels on nodes are deprecated and will be removed in a future release. You can follow the list below to see which labels are being replaced:

Please note: The following changes does not have a set Kubernetes release. However, the replacement labels are already implemented.

• beta.kubernetes.io/instance-type -> node.kubernetes.io/instance-type
• beta.kubernetes.io/arch -> kubernetes.io/arch
• beta.kubernetes.io/os -> kubernetes.io/os
• failure-domain.beta.kubernetes.io/region -> topology.kubernetes.io/region
• failure-domain.beta.kubernetes.io/zone -> topology.kubernetes.io/zone

Is downtime expected

The upgrade drains (move all workload from) one node at the time, patches that node and brings it back in the cluster. First after all deployments and statefulsets are running again we will continue on with the next node.

Known issues

Custom node taints and labels lost during upgrade

All custom taints and labels on nodes are lost during upgrade.

Custom changes to non -customer security groups will be lost

All changes to security groups not suffixed with “-customer” will be lost during the upgrade.

Snapshots are not working

There is currently a limitation in the snapshot controller not making it topology aware.