1. Create an NBSetupKey object referring to your secret.
```yaml
apiVersion:netbird.io/v1
kind:NBSetupKey
metadata:
name:test
spec:
# Optional, overrides management URL for this setupkey only
# defaults to https://api.netbird.io
managementURL:https://netbird.example.com
secretKeyRef:
name:test# Required
key:setupkey# Required
```
1. Annotate the pods you need to inject NetBird into with `netbird.io/setup-key`.
```yaml
kind:Deployment
...
spec:
...
template:
metadata:
annotations:
netbird.io/setup-key:test# Must match the name of an NBSetupKey object in the same namespace
...
spec:
containers:
...
```
Since v0.27.0, NetBird supports extra DNS labels, which extends the DNS names that you can link to peers by grouping them and load balancing access using DNS round-robin. To enable this feature, add the following annotation to the pod:
```yaml
netbird.io/extra-dns-labels:"label1,label2"
```
With this setup, all peers with the same extra label would be used in a DNS round-robin fashion.
By default, the NetBird container is injected as a regular sidecar container. For workloads like Jobs and CronJobs where the pod should terminate after the main container completes, you can use init sidecar mode. This injects NetBird as an init container with `restartPolicy: Always`.
To enable init sidecar mode, add the following annotation:
### Granting controller access to NetBird Management
> [!IMPORTANT]
> The NetBird Kubernetes operator generates configurations using NetBird API; editing or deleting these configurations in the NetBird console may cause temporary network disconnection until the operator reconciles the configuration.
1. Create a Service User on your NetBird dashboard (Must be Admin). [Doc](https://docs.netbird.io/how-to/access-netbird-public-api#creating-a-service-user).
2. Create an access token for the Service User (Must be Admin). [Doc](https://docs.netbird.io/how-to/access-netbird-public-api#creating-a-service-user).
3. Add access token to your helm values file under `netbirdAPI.key`.
3. Set `ingress.kubernetesAPI.groups` to a list of groups to assign to the Network Resource to be created for kubernetes API.
4. Set `ingress.kubernetesAPI.policies` to a list of policy names to connect to the resource (See #managing-policies for more details).
5. Apply Helm changes through `helm upgrade`.
6. Replace the server URL in your kubeconfig file with `https://kubernetes.default.<cluster DNS>` (by default `https://kubernetes.default.svc.cluster.local`), for example:
> Ingress DNS Resolution requires DNS Wildcard Routing to be enabled and at least one DNS Nameserver configured for clients.
|Annotation|Description|Default|Valid Values|
|---|---|---|---|
|`netbird.io/expose`| Expose service using NetBird Network Resource ||(`null`, `true`)|
|`netbird.io/groups`| Comma-separated list of group names to assign to Network Resource |`{ClusterName}-{Namespace}-{Service}`|Any comma-separated list of strings.|
|`netbird.io/resource-name`| Network Resource name |`{Namespace}-{Service}`|Any valid network resource name, make sure they're unique!|
|`netbird.io/policy-ports`| Narrow down exposed ports in a policy. Leave empty for all ports. ||Comma-separated integer list, integers must be between 0-65535|
|`netbird.io/policy-protocol`| Narrow down protocol for use in a policy. Leave empty for all protocols. ||(`tcp`,`udp`)|
|`netbird.io/policy-source-groups`| Specify source groups for auto-generated policies. Required for auto-generating policies||Any comma-separated list of strings.|
|`netbird.io/policy-name`| Specify human-friendly names for auto-generated policies. ||comma-separated list of `policy:friendly-name`, where policy is the name of the kubernetes object.|
*`netbird.io/expose` will interpret any string as a `true` value; the only `false` value is `null`.
* The operator does **not** handle duplicate resource names within the same network, it is up to you to ensure resource names are unique within the same network.
* While the NetBird console will allow group names to contain commas, this is not allowed in `netbird.io/groups` annotation as commas are used as separators.
* If a group already exists on NetBird console with the same name, NetBird Operator will use that group ID instead of creating a new group.
* NetBird Operator will attempt to clean up any resources created, including groups created for resources.
* If a group is used by resources that the operator cannot clean up, the operator will eventually ignore the group in NetBird.
* It's recommended that unique groups be used per NetBird Operator installation to remove any possible conflicts.
* The Operator does not validate service annotations on updates, as this may cause unnecessary overhead on any Service update.
1. Ensure `ingress.allowAutomaticPolicyCreation` is set to true in the Helm chart and apply.
2. Annotate a service with `netbird.io/policy` with the name of the policy as a kubernetes object, for example `netbird.io/policy: default`. This will create an NBPolicy with the name `default-<Service Namespace>-<Service Name>`.
3. Annotate the same service with `netbird.io/policy-source-groups` with a comma-separated list of group names to allow as a source, for example `netbird.io/policy-source-groups: dev`.
4. (Optional) Annotate the service with `netbird.io/policy-name` for a human-friendly name, for example `netbird.io/policy-name: "default:Default policy for kubernetes cluster"`.
* Each NBPolicy will only create policies in the NetBird console when the information provided is enough to create one. If no services act as a destination or specified services do not conform to the protocol(s) defined, the policy will not be created.
* Each NBPolicy will create one policy in the NetBird console per protocol specified as long as the protocol has destinations; this ensures better-secured policies by separating ports for TCP and UDP.
* Policies currently do not support ICMP protocol, as ICMP is not supported in Kubernetes services, and there are [no current plans to support it](https://discuss.kubernetes.io/t/icmp-support-for-kubernetes-service/21738).