Multus

Multus is a container network interface (CNI) plugin that enables the attachment of multiple network interfaces to a single Pod.

By default, Pods in Kubernetes are connected to a single network interface, which is the default network. With Multus CNI, though, Pods can have multiple network interfaces for multi-homed connectivity. For more information, refer to the Multus CNI GitHub repository.

Enable Multus

Multus is disabled in MKE 4 by default. To enable the function, you must obtain the MKE 4 configuration file, locate the network.multus.enabled section, set the enabled parameter to true, and apply the new configuration.

Obtain the default MKE 4 configuration file:
```
mkectl init
```
Navigate to the network section of the configuration file, and set the enabled parameter for multus to true.
```
network:
  multus:
    enabled: true
```

Apply the configuration:

mkectl apply -f <mke-configuration-file>

Verify the successful deployment of MetalLB in the cluster:

kubectl get daemonset,pods -n kube-system -l app=multus

Example output:

NAME                            DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR   AGE
daemonset.apps/kube-multus-ds   3         3         3       3            3           <none>          50s

NAME                       READY   STATUS    RESTARTS   AGE
pod/kube-multus-ds-8psck   1/1     Running   0          36s
pod/kube-multus-ds-dltjh   1/1     Running   0          36s
pod/kube-multus-ds-m2bsz   1/1     Running   0          36s

In MKE 4, you can disable Multus at any time, as opposed to MKE 3 where once Multus is installed it cannot be disabled.

Add a network interface

Run the following command on all of the nodes in the cluster to download and extract the CNI plugin:

CNI_PLUGIN_VERSION=v1.3.0
CNI_ARCH=amd64
curl -sL
https://github.com/containernetworking/plugins/releases/download/${CNI_PLUGIN_VERSION}/cni-plugins-linux-${CNI_ARCH}-${CNI_PLUGIN_VERSION}.tgz
| sudo tar xvz -C /opt/cni/bin/

Determine the primary network interface for the node. You will use this information to create the NetworkAttachmentDefinitions file.

ℹ️

The name of the primary interface can vary with the underlying network adapter.

route

ℹ️

eth0 is the primary network interface for most Linux distributions.

Sample output:

Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
default         ip-172-31-0-1.u 0.0.0.0         UG    100    0        0 ens5
172.31.0.0      0.0.0.0         255.255.0.0     U     100    0        0 ens5
ip-172-31-0-1.u 0.0.0.0         255.255.255.255 UH    100    0        0 ens5
192.168.17.0    0.0.0.0         255.255.255.192 U     0      0        0 *

Create the NetworkAttachmentDefinitions file, to specify other networks:

cat <<EOF | kubectl create -f -
apiVersion: k8s.cni.cncf.io/v1
kind: NetworkAttachmentDefinition
metadata:
  name: ens5-network
spec:
  config: |
    {
      "cniVersion": "0.3.1",
      "type": "macvlan",
      "master": "ens5",
      "mode": "bridge",
      "mtu": 9001,
      "ipam": {
        "type": "host-local",
        "subnet": "172.31.0.0/16",
        "rangeStart": "172.31.2.150",
        "rangeEnd": "172.31.2.200",
        "routes": [
          { "dst": "0.0.0.0/0" }
        ],
        "gateway": "172.31.2.1"
      }
    }
EOF

Verify the creation of the the network attachment definition:

kubectl get network-attachment-definition
NAME           AGE
ens5-network   44s

Create a multi-homed Pod:

cat <<EOF | kubectl create -f -
apiVersion: v1
kind: Pod
metadata:
  name: pod-additional-network
  annotations:
    k8s.v1.cni.cncf.io/networks: ens5-network
spec:
  containers:
    - command:
        - sleep
        - "3600"
      image: busybox
      name: pods-simple-container
EOF

Verify the network interfaces of the Pod:

kubectl exec -it pod-additional-network -- ip a

Sample output:

LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0@if15: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 8951 qdisc noqueue qlen 1000
    link/ether 26:36:4c:44:9c:80 brd ff:ff:ff:ff:ff:ff
    inet 192.168.23.138/32 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::2436:4cff:fe44:9c80/64 scope link
       valid_lft forever preferred_lft forever
3: net1@eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9001 qdisc noqueue
    link/ether 0e:a3:f7:e8:50:85 brd ff:ff:ff:ff:ff:ff
    inet 172.31.2.150/16 brd 172.31.255.255 scope global net1
       valid_lft forever preferred_lft forever
    inet6 fe80::ca3:f7ff:fee8:5085/64 scope link
       valid_lft forever preferred_lft forever

Uninstall Multus

Obtain the MKE 4 configuration file.
Set the enabled field to false to disable Multus.
```
network:
  multus:
    enabled: false
```

Apply the configuration:

mkectl apply -f <mke-configuration-file>

Last updated on February 26, 2025

Workload node deployment