feature, and can be performed by different people/teams at different times.

It is strongly recommended that different namespaces are used, with different service accounts.
In particular access to the security-critical production components (root CA, policy, control)
should be locked down and restricted.  The new installer allows multiple instances of

// (root certificates).
//
// ClusterTrustBundle objects are considered to be readable by any authenticated
// user in the cluster, because they can be mounted by pods using the
// `clusterTrustBundle` projection.  All service accounts have read access to
// ClusterTrustBundles by default.  Users who only have namespace-level access
// to a cluster can read ClusterTrustBundles by impersonating a serviceaccount
// that they have access to.
//

  // +optional
  map<string, string> nodeSelector = 7;

  // ServiceAccountName is the name of the ServiceAccount to use to run this carp.
  // More info: https://kubernetes.io/docs/concepts/security/service-accounts/
  // +optional
  optional string serviceAccountName = 8;

  // DeprecatedServiceAccount is a depreciated alias for ServiceAccountName.
  // Deprecated: Use serviceAccountName instead.

ISTIO_META_MESH_ID=''
ISTIO_META_NETWORK=''
ISTIO_META_WORKLOAD_NAME='foo'
ISTIO_NAMESPACE='bar'
ISTIO_SERVICE='foo.bar'
ISTIO_SERVICE_CIDR='*'
ISTIO_SVC_IP='10.10.10.10'
POD_NAMESPACE='bar'
SERVICE_ACCOUNT='vm-serviceaccount'

    - creates service-account `istio-cni` with `ClusterRoleBinding` to allow gets on pods' info and delete/modifications for recovery.

- `install-cni` container
    - copies `istio-cni` and `istio-iptables` to `/opt/cni/bin`
    - creates kubeconfig for the service account the pod runs under
    - periodically copy the K8S JWT token for istio-cni on the host to connect to K8S.

  // Allowing surge implies the possibility that the resources consumed by the
  // daemonset on any given node can double if the readiness check fails, and
  // so resource intensive daemonsets should take into account that they may
  // cause evictions during disruption.
  // +optional
  optional k8s.io.apimachinery.pkg.util.intstr.IntOrString maxSurge = 2;
}

// Spec to control the desired behavior of rolling update.

  // uncountedTerminatedPods holds the UIDs of Pods that have terminated but
  // the job controller hasn't yet accounted for in the status counters.
  //
  // The job controller creates pods with a finalizer. When a pod terminates
  // (succeeded or failed), the controller does three steps to account for it
  // in the job status:
  //
  // 1. Add the pod UID to the arrays in this field.
  // 2. Remove the pod finalizer.

ISTIO_META_NETWORK=''
ISTIO_META_WORKLOAD_NAME='foo'
ISTIO_NAMESPACE='bar'
ISTIO_SERVICE='foo.bar'
ISTIO_SERVICE_CIDR='*'
ISTIO_SVC_IP='fd00:10:96::1'
POD_NAMESPACE='bar'
PROXY_CONFIG_ANNOT_VALUE='bar'
SERVICE_ACCOUNT='vm-serviceaccount'

```

Ensure the created binary is in your PATH to run the examples below.

### Controller (in cluster)

Building a custom controller requires a Dockerhub (or similar) account. To build using the container based build:

```bash
HUB=docker.io/<your-account> TAG=latest make docker.operator
```

This builds the controller binary and docker file, and pushes the image to the specified hub with the `latest` tag.

However, the identities of the certificates will be of the actual user workloads, not Ztunnel's own identity.
This means Ztunnel will have multiple distinct certificates at a time, one for each unique identity (service account) running on its node.

When fetching certificates, ztunnel will authenticate to the CA with its own identity, but request the identity of another workload.