fb557ab20f
DPDeviceFile is used by Device Plugins to pass device data to CNIs, as defined in the device-info-spec. The name of the DPDeviceFile is defined by the device-info-spec as: <ResourceName>-<DeviceID>-device.json If the DPDeviceFile exists, the NPWG implementation makes a copy of the file and passes the name of the file to the delegate CNI via capabilityArgs as CNIDeviceFile. If the DPDeviceFile does not exist, the filename is still passed to the CNI. The CNI can create the file and popluate it if a device is created within the CNI. The name of the CNIDeviceFile is not defined by device-info-spec, but to ensure the name does not clash it is formed by the following unique triplet: [networkName, PodUUID, ifName] k8snetworkplumbingwg/network-attachment-definition-client repo has utility functions to abstract some of this functionality so it can be reused across Device Plugins, NPWG implementations and CNIs. Signed-off-by: Billy McFall <22157057+Billy99@users.noreply.github.com> |
||
|---|---|---|
| .github | ||
| checkpoint | ||
| doc | ||
| e2e | ||
| examples | ||
| images | ||
| k8sclient | ||
| kubeletclient | ||
| logging | ||
| multus | ||
| netutils | ||
| testing | ||
| types | ||
| vendor | ||
| .gitignore | ||
| .goreleaser.yml | ||
| build | ||
| CODE_OF_CONDUCT.md | ||
| CONTRIBUTING.md | ||
| Dockerfile | ||
| Dockerfile.arm64 | ||
| Dockerfile.openshift | ||
| Dockerfile.ppc64le | ||
| go.mod | ||
| go.sum | ||
| LICENSE | ||
| README.md | ||
| test.sh | ||
Multus-CNI
Multus CNI enables attaching multiple network interfaces to pods in Kubernetes.
How it works
Multus CNI is a container network interface (CNI) plugin for Kubernetes that enables attaching multiple network interfaces to pods. Typically, in Kubernetes each pod only has one network interface (apart from a loopback) -- with Multus you can create a multi-homed pod that has multiple interfaces. This is accomplished by Multus acting as a "meta-plugin", a CNI plugin that can call multiple other CNI plugins.
Multus CNI follows the Kubernetes Network Custom Resource Definition De-facto Standard to provide a standardized method by which to specify the configurations for additional network interfaces. This standard is put forward by the Kubernetes Network Plumbing Working Group.
Multus is one of the projects in the Baremetal Container Experience kit
Multi-Homed pod
Here's an illustration of the network interfaces attached to a pod, as provisioned by Multus CNI. The diagram shows the pod with three interfaces: eth0, net0 and net1. eth0 connects kubernetes cluster network to connect with kubernetes server/services (e.g. kubernetes api-server, kubelet and so on). net0 and net1 are additional network attachments and connect to other networks by using other CNI plugins (e.g. vlan/vxlan/ptp).
Quickstart Installation Guide
The quickstart installation method for Multus requires that you have first installed a Kubernetes CNI plugin to serve as your pod-to-pod network, which we refer to as your "default network" (a network interface that every pod will be creatd with). Each network attachment created by Multus will be in addition to this default network interface. For more detail on installing a default network CNI plugins, refer to our quick-start guide.
Clone this GitHub repository, we'll apply a daemonset which installs Multus using to kubectl from this repo. From the root directory of the clone, apply the daemonset YAML file:
$ cat ./images/multus-daemonset.yml | kubectl apply -f -
This will configure your systems to be ready to use Multus CNI, but, to get started with adding additional interfaces to your pods, refer to our complete quick-start guide
Additional installation Options
- Install via daemonset using the quick-start guide, above.
- Download binaries from release page
- By Docker image from Docker Hub
- Or, roll-your-own and build from source
- See Development
Comprehensive Documentation
Contact Us
For any questions about Multus CNI, feel free to ask a question in #general in the Intel-Corp Slack, or open up a GitHub issue.