Example: Installing Calico and Setting Up Network Policies
Find out how to install Calico and set up network policies on a cluster you've created using Kubernetes Engine (OKE).
The Kubernetes networking model assumes containers (pods) have unique and routable IP addresses within a cluster. In the Kubernetes networking model, containers communicate with each other using those IP addresses, regardless of whether the containers are deployed on the same node in a cluster or on a different node. Kubernetes has adopted the Container Network Interface (CNI) specification for network resource management. The CNI consists of a specification and libraries for writing plugins to configure network interfaces in Linux containers, along with a number of supported plugins.
By default, pods accept traffic from any source. To enhance cluster security, pods can be 'isolated' by selecting them in a network policy (the Kubernetes NetworkPolicy resource). A network policy is a specification of how groups of pods are allowed to communicate with each other and other network endpoints. NetworkPolicy resources use labels to select pods and to define rules that specify what traffic is allowed to the selected pods. If a NetworkPolicy in a cluster namespace selects a particular pod, that pod will reject any connections that are not allowed by any NetworkPolicy. Other pods in the namespace that are not selected by a NetworkPolicy will continue to accept all traffic. For more information about network policies, see the Kubernetes documentation.
Network policies are implemented by the CNI network provider. Simply creating the NetworkPolicy resource without a CNI network provider to implement it will have no effect. Note that not all CNI network providers implement the NetworkPolicy resource.
When you create clusters with Kubernetes Engine, you select a Network type. The Network type you select determines the CNI network provider and associated CNI plugin that is used for pod networking, as follows:
- VCN-native pod networking: Uses the OCI VCN-Native Pod Networking CNI plugin to connect worker nodes to pod subnets in an Oracle Cloud Infrastructure VCN. As a result, pod IP addresses within a VCN are directly routable from other VCNs connected (peered) to that VCN, from on-premise networks, and from the internet. See Using the OCI VCN-Native Pod Networking CNI plugin for pod networking.
- Flannel overlay: Uses the flannel CNI plugin to encapsulate communication between pods in the flannel overlay network, a simple private overlay virtual network that attaches IP addresses to containers. The pods in the private overlay network are only accessible from other pods in the same cluster. See Using the flannel CNI plugin for pod networking.
Although both the OCI VCN-Native Pod Networking CNI plugin and the flannel CNI plugin satisfy the requirements of the Kubernetes networking model, neither of them support Kubernetes NetworkPolicy resources. If you want to enhance the security of clusters you create with Kubernetes Engine by implementing network policies, you have to install and configure a network provider that does support NetworkPolicy resources alongside the CNI plugin used in the cluster. One such provider is Calico (refer to the Kubernetes documentation for a list of other network providers).
Calico is an open source networking and network security solution for containers, virtual machines, and native host-based workloads. For more information about Calico, see the Calico documentation.
- You can use Calico with managed node pools, but not with virtual node pools.
-
Only the use of open source Calico is supported. Use of Calico Enterprise is not supported.
- In the case of flannel, if you install Calico on a cluster that has existing node pools in which pods are already running, you will have to restart the pods when the Calico installation is complete. For example, by running the
kubectl rollout restart
command. If you install Calico on a cluster before creating any node pools in the cluster (recommended), you can be sure that there will be no pods to restart.
Calico Compatibility
The table lists the versions of the Calico network plugin that Oracle has successfully tested on clusters created using Kubernetes Engine. Oracle only supports Calico versions that have been successfully tested. For each Calico version, the table shows the Kubernetes version that was running on clusters in successful tests.
Note that only the use of open source Calico is supported. Use of Calico Enterprise is not supported.
For more information, see Example: Installing Calico and Setting Up Network Policies.
Calico Version | Tested (and supported) on clusters running Kubernetes 1.28? | Tested (and supported) on clusters running Kubernetes 1.29? | Tested (and supported) on clusters running Kubernetes 1.30? | Tested (and supported) on clusters running Kubernetes 1.31? |
---|---|---|---|---|
3.25.1 | (not tested) | (not tested) | (not tested) | (not tested) |
3.26.1 | (not tested) | (not tested) | (not tested) | (not tested) |
3.26.4 | Yes | (not tested) | (not tested) | (not tested) |
3.27.2 | (not tested) | Yes | (not tested) | (not tested) |
3.28.0 | (not tested) | (not tested) | Yes | (not tested) |
3.28.2 | (not tested) | (not tested) | (not tested) | Yes |
Installing Calico alongside the flannel CNI plugin
Having created a cluster using Kubernetes Engine (using either the Console or the API) and selected flannel overlay as the Network type, you can subsequently install Calico on the cluster alongside the flannel CNI plugin to support network policies.
For convenience, Calico installation instructions are included below. Note that Calico installation instructions vary between Calico versions. For information about installing different versions of Calico, always refer to the Calico installation documentation.
-
If you haven't already done so, follow the steps to set up the cluster's kubeconfig configuration file and (if necessary) set the KUBECONFIG environment variable to point to the file. Note that you must set up your own kubeconfig file. You cannot access a cluster using a kubeconfig file that a different user set up. See Setting Up Cluster Access.
-
In a terminal window, download the Calico policy-only manifest for the Kubernetes API datastore by entering:
curl https://raw.githubusercontent.com/projectcalico/calico/v3.25.1/manifests/calico-policy-only.yaml -o calico.yaml
Note that the url differs, according to the version of Calico that you want to install.
- You have to set one or more additional environment variables in the calico-policy-only.yaml file.
- Open the calico-policy-only.yaml file in a text editor of your choice.
- Add the following environment variables for the
calico-node
container in the manifest of thecalico-node
DaemonSet:CALICO_MANAGE_CNI="true"
FELIX_IPTABLESBACKEND="NFT"
Note: Only add this environment variable if you have selected an Oracle Linux 8 image for worker nodes in the cluster.
Before you make the change, the
calico-node
container environment variables (env:
) section of thecalico-node
DaemonSet manifest looks like this:... containers: # Runs calico-node container on each Kubernetes node. This # container programs network policy and routes on each # host. - name: calico-node image: docker.io/calico/node:v3.25.1 env: # Use Kubernetes API as the backing datastore. - name: DATASTORE_TYPE value: "kubernetes" # Wait for the datastore. - name: WAIT_FOR_DATASTORE value: "true" # Set based on the k8s node name. - name: NODENAME valueFrom: fieldRef: fieldPath: spec.nodeName # Don't enable BGP. - name: CALICO_NETWORKING_BACKEND value: "none" - name: CLUSTER_TYPE value: "k8s" - name: FELIX_HEALTHENABLED value: "true" - name: FELIX_LOGSEVERITYSCREEN value: "Info" # Disable file logging so `kubectl logs` works. - name: CALICO_DISABLE_FILE_LOGGING value: "true" # Disable IPv6 on Kubernetes. - name: FELIX_IPV6SUPPORT value: "false"
After you have made the change, check that the
calico-node
container environment variables (env:
) section of thecalico-node
DaemonSet manifest looks like this:... containers: # Runs calico-node container on each Kubernetes node. This # container programs network policy and routes on each # host. - name: calico-node image: docker.io/calico/node:v3.25.1 env: # Use Kubernetes API as the backing datastore. - name: DATASTORE_TYPE value: "kubernetes" # Wait for the datastore. - name: WAIT_FOR_DATASTORE value: "true" # Set based on the k8s node name. - name: NODENAME valueFrom: fieldRef: fieldPath: spec.nodeName # Don't enable BGP. - name: CALICO_NETWORKING_BACKEND value: "none" - name: CLUSTER_TYPE value: "k8s" - name: FELIX_HEALTHENABLED value: "true" - name: FELIX_LOGSEVERITYSCREEN value: "Info" # Disable file logging so `kubectl logs` works. - name: CALICO_DISABLE_FILE_LOGGING value: "true" # Disable IPv6 on Kubernetes. - name: FELIX_IPV6SUPPORT value: "false" - name: CALICO_MANAGE_CNI value: "true" - name: FELIX_IPTABLESBACKEND value: "NFT" ...
- Save and close the modified
calico-policy-only.yaml
file.
-
Install and configure Calico by entering the following command:
kubectl apply -f calico.yaml
Installing Calico alongside the OCI VCN-Native Pod Networking CNI plugin
Having created a cluster using Kubernetes Engine (using either the Console or the API) and selected VCN-native pod networking as the Network type, you can subsequently install Calico on the cluster alongside the OCI VCN-Native Pod Networking CNI plugin to support network policies.
For convenience, Calico installation instructions are included below. Note that Calico installation instructions vary between Calico versions. For information about installing different versions of Calico, always refer to the Calico installation documentation.
-
If you haven't already done so, follow the steps to set up the cluster's kubeconfig configuration file and (if necessary) set the KUBECONFIG environment variable to point to the file. Note that you must set up your own kubeconfig file. You cannot access a cluster using a kubeconfig file that a different user set up. See Setting Up Cluster Access.
-
In a terminal window, download the Calico policy-only manifest for the Kubernetes API datastore by entering:
curl https://raw.githubusercontent.com/projectcalico/calico/v3.25.1/manifests/calico-policy-only.yaml -o calico.yaml
Note that the url differs, according to the version of Calico that you want to install.
- The calico-policy-only.yaml file includes Calico components that are not required when using Calico alongside the OCI VCN-Native Pod Networking CNI plugin, so you have to remove these components. You also have to set some additional environment variables.
- Open the calico-policy-only.yaml file in a text editor of your choice.
- Remove the
initContainers
section from the manifest of thecalico-node
DaemonSet. - Remove the following from the
env
section for thecalico-node
container from the manifest of thecalico-node
DaemonSet:# Typha support: controlled by the ConfigMap. - name: FELIX_TYPHAK8SSERVICENAME valueFrom: configMapKeyRef: name: calico-config key: typha_service_name
- Remove the following
envFrom
section for thecalico-node
container from the manifest of thecalico-node
DaemonSet:envFrom: - configMapRef: # Allow KUBERNETES_SERVICE_HOST and KUBERNETES_SERVICE_PORT to be overridden for eBPF mode. name: kubernetes-services-endpoint optional: true
- Remove the following volumes from the
volumes
section of the manifest of thecalico-node
DaemonSet:cni-bin-dir
cni-net-dir
cni-log-dir
Before you make the change, the
volumes
section of thecalico-node
DaemonSet manifest looks like this:... volumes: # Used by calico-node. - name: lib-modules hostPath: path: /lib/modules - name: var-run-calico hostPath: path: /var/run/calico - name: var-lib-calico hostPath: path: /var/lib/calico - name: xtables-lock hostPath: path: /run/xtables.lock type: FileOrCreate - name: sysfs hostPath: path: /sys/fs/ type: DirectoryOrCreate # Used to install CNI. - name: cni-bin-dir hostPath: path: /opt/cni/bin - name: cni-net-dir hostPath: path: /etc/cni/net.d # Used to access CNI logs. - name: cni-log-dir hostPath: path: /var/log/calico/cni # Used to create per-pod Unix Domain Sockets - name: policysync hostPath: type: DirectoryOrCreate path: /var/run/nodeagent ...
After you have made the change, check that the
volumes
section of thecalico-node
DaemonSet manifest looks like this:... volumes: # Used by calico-node. - name: lib-modules hostPath: path: /lib/modules - name: var-run-calico hostPath: path: /var/run/calico - name: var-lib-calico hostPath: path: /var/lib/calico - name: xtables-lock hostPath: path: /run/xtables.lock type: FileOrCreate - name: sysfs hostPath: path: /sys/fs/ type: DirectoryOrCreate # Used to create per-pod Unix Domain Sockets - name: policysync hostPath: type: DirectoryOrCreate path: /var/run/nodeagent ...
- Remove the following volume mounts from the
volumeMounts
section for thecalico-node
container in the manifest of thecalico-node
DaemonSet:cni-net-dir
, including the associated comment# For maintaining CNI plugin API credentials.
cni-log-dir
Before you make the change, the
volumeMounts
section looks like this:... volumeMounts: # For maintaining CNI plugin API credentials. - mountPath: /host/etc/cni/net.d name: cni-net-dir readOnly: false - mountPath: /lib/modules name: lib-modules readOnly: true - mountPath: /run/xtables.lock name: xtables-lock readOnly: false - mountPath: /var/run/calico name: var-run-calico readOnly: false - mountPath: /var/lib/calico name: var-lib-calico readOnly: false - name: policysync mountPath: /var/run/nodeagent # For eBPF mode, we need to be able to mount the BPF filesystem at /sys/fs/bpf so we mount in the # parent directory. - name: bpffs mountPath: /sys/fs/bpf - name: cni-log-dir mountPath: /var/log/calico/cni readOnly: true ...
After you have made the change, check that the
volumeMounts
section looks like this:... volumeMounts: - mountPath: /lib/modules name: lib-modules readOnly: true - mountPath: /run/xtables.lock name: xtables-lock readOnly: false - mountPath: /var/run/calico name: var-run-calico readOnly: false - mountPath: /var/lib/calico name: var-lib-calico readOnly: false - name: policysync mountPath: /var/run/nodeagent # For eBPF mode, we need to be able to mount the BPF filesystem at /sys/fs/bpf so we mount in the # parent directory. - name: bpffs mountPath: /sys/fs/bpf ...
- Add the following environment variables for the
calico-node
container in the manifest of thecalico-node
DaemonSet:FELIX_INTERFACEPREFIX="oci"
NO_DEFAULT_POOLS="true"
FELIX_CHAININSERTMODE="Append"
FELIX_IPTABLESMANGLEALLOWACTION="Return"
FELIX_IPTABLESBACKEND="NFT"
Note: Only add this environment variable if you have selected an Oracle Linux 8 image for worker nodes in the cluster.
Before you make the change, the
calico-node
container environment variables (env:
) section of thecalico-node
DaemonSet manifest looks like this:... containers: # Runs calico-node container on each Kubernetes node. This # container programs network policy and routes on each # host. - name: calico-node image: docker.io/calico/node:v3.25.1 env: # Use Kubernetes API as the backing datastore. - name: DATASTORE_TYPE value: "kubernetes" # Wait for the datastore. - name: WAIT_FOR_DATASTORE value: "true" # Set based on the k8s node name. - name: NODENAME valueFrom: fieldRef: fieldPath: spec.nodeName # Don't enable BGP. - name: CALICO_NETWORKING_BACKEND value: "none" - name: CLUSTER_TYPE value: "k8s" - name: FELIX_HEALTHENABLED value: "true" - name: FELIX_LOGSEVERITYSCREEN value: "Info" # Disable file logging so `kubectl logs` works. - name: CALICO_DISABLE_FILE_LOGGING value: "true" # Disable IPv6 on Kubernetes. - name: FELIX_IPV6SUPPORT value: "false"
After you have made the change, check that the
calico-node
container environment variables (env:
) section of thecalico-node
DaemonSet manifest looks like this:... containers: # Runs calico-node container on each Kubernetes node. This # container programs network policy and routes on each # host. - name: calico-node image: docker.io/calico/node:v3.25.1 env: # Use Kubernetes API as the backing datastore. - name: DATASTORE_TYPE value: "kubernetes" # Wait for the datastore. - name: WAIT_FOR_DATASTORE value: "true" # Set based on the k8s node name. - name: NODENAME valueFrom: fieldRef: fieldPath: spec.nodeName # Don't enable BGP. - name: CALICO_NETWORKING_BACKEND value: "none" - name: CLUSTER_TYPE value: "k8s" - name: FELIX_HEALTHENABLED value: "true" - name: FELIX_LOGSEVERITYSCREEN value: "Info" # Disable file logging so `kubectl logs` works. - name: CALICO_DISABLE_FILE_LOGGING value: "true" # Disable IPv6 on Kubernetes. - name: FELIX_IPV6SUPPORT value: "false" # Configuration for Native Pod Networking - name: FELIX_INTERFACEPREFIX value: "oci" - name: NO_DEFAULT_POOLS value: "true" - name: FELIX_CHAININSERTMODE value: "Append" - name: FELIX_IPTABLESMANGLEALLOWACTION value: "Return" - name: FELIX_IPTABLESBACKEND value: "NFT" ...
- Save and close the modified
calico-policy-only.yaml
file.
-
Install and configure Calico by entering the following command:
kubectl apply -f calico.yaml
Setting up Network Policies
Having installed Calico on a cluster you've created with Kubernetes Engine, you can create Kubernetes NetworkPolicy resources to isolate pods as required.
For NetworkPolicy examples and how to use them, see the Calico documentation and specifically:
Note that the examples vary, according to the Calico version you've installed.