Tailscale Kubernetes operator v1.84.0

A new release of the Tailscale Kubernetes operator is available. For guidance on installing and updating, refer to our installation instructions.

• New: Tailscale Ingress resource supports high availability (HA) mode and multiplexing by using a ProxyGroup. You can expose an Ingress resource to a tailnet by using multiple active proxy replicas (Pods). You can multiplex multiple Ingress resources on the same set of proxy Pods.

• New: Tailscale Kubernetes Services support HA mode and multiplexing. You can expose a cluster app to a tailnet by using multiple active network layer proxy Pods to help prevent downtime. You can expose multiple apps to a tailnet on the same set of proxy Pods.

• New: Tailscale Ingress supports exposing applications deployed across multiple clusters (multi-cluster Ingress) to the tailnet.

• New: Pods deployed for a Recorder resource can use AWS IAM Roles for Service Accounts (IRSA) instead of static Amazon S3 credentials by configuring the created ServiceAccount object's name and annotations.

• New: Tailscale Kubernetes Services support exposing to tailnet applications that are deployed across multiple clusters (multi-cluster Service).

• Changed: Tailscale Kubernetes operator needs to watch EndpointSlice objects at cluster scope, to ensure failover for multi-cluster Service and Ingress resources in cases where there are no healthy backends in one of the clusters.

• Fixed: The Kubernetes Operator will default any path left unset on an Ingress resource to the / path.