README
¶
Accelerated Bridge CNI plugin
This plugin allows using Linux Bridge with hardware offloading in containers and orchestrators like Kubernetes. Accelerated Bridge CNI plugin requires NIC with support of SR-IOV technology with VFs in switchdev mode and support of Linux Bridge offloading.
In switchdev mode, two net-devs exist for VF net-device and VF representor. A packet sent through the VF representor on the host arrives to the VF, and a packet sent through the VF is received by its representor. CNI plugin moves VF net-device to a container network namespace and attaches VF representor to a Linux Bridge on a host. Then additional Bridge port settings, such as VLANs configuration, can be applied for VF representor.
This plugin works with SR-IOV device plugin for VF allocation in Kubernetes.
A metaplugin such as Multus gets the allocated VF's deviceID(PCI address) and is responsible for invoking the Accelerated Bridge CNI plugin with that deviceID.
Accelerated Bridge plugin assumes that Linux Bridge is already exist and correctly configured on nodes.
Accelerated bridge CNI supports automatic Linux bridge selection if multiple bridges are set in the configuration. The Plugin checks to which Linux bridge uplink for VF is attached and uses that bridge to add a VF representor. Automatic bridge selection logic requires uplink to be added to a bridge before calling the CNI.
Supported configurations for auto bridge selection:
- uplink is a direct member of a Linux bridge
- uplink is a part of a bond interface, bond interface is a member of a Linux bridge
CNI plugin also supports VF bound to userspace driver (currently only vfio-pci) which may be utilized for virtualization use-case i.e KubeVirt. If CNI plugin detects that VF bounded to a userspace driver, it will skip step with VF netdev configuration.
Build
This plugin uses Go modules for dependency management and requires Go 1.18 to build.
To build the plugin binary:
make
Upon successful build the plugin binary will be available in build/accelerated-bridge.
Kubernetes Quick Start
A full guide on orchestrating SR-IOV virtual functions in Kubernetes can be found at the SR-IOV Network Device Plugin project.
Creating VFs is outside the scope of the Accelerated Bridge CNI plugin. More information about allocating VFs on different NICs can be found here
To deploy Accelerated Bridge CNI by itself on a Kubernetes 1.16+ cluster:
kubectl apply -f images/k8s-v1.16/accelerated-bridge-cni-daemonset.yaml
Usage
Accelerated Bridge CNI networks are commonly configured using Multus and SR-IOV Device Plugin using Network Attachment Definitions. More information about configuring Kubernetes networks using this pattern can be found in the Multus configuration reference document.
A Network Attachment Definition for Accelerated Bridge CNI takes the form:
apiVersion: "k8s.cni.cncf.io/v1"
kind: NetworkAttachmentDefinition
metadata:
name: some-net
annotations:
k8s.v1.cni.cncf.io/resourceName: intel.com/intel_sriov_netdevice
spec:
config: '{
"type": "accelerated-bridge",
"cniVersion": "0.3.1",
"name": "some-net",
"capabilities": {"CNIDeviceInfoFile": true},
"ipam": {
"type": "host-local",
"subnet": "10.56.217.0/24",
"routes": [{
"dst": "0.0.0.0/0"
}],
"gateway": "10.56.217.1"
}
}'
The .spec.config field contains the configuration information used by the Accelerated Bridge CNI.
Note: "capabilities": {"CNIDeviceInfoFile": true} is required if you want Multus to pass on
DeviceInfo file
to the plugin. Accelerated Bridge plugin will add additional PCI device information to the file if the file is provided.
Basic configuration parameters
The following parameters are generic parameters which are not specific to the Accelerated Bridge CNI configuration, though (with the exception of ipam) they need to be included in the config.
cniVersion: the version of the CNI spec used.type: CNI plugin used. "accelerated-bridge" corresponds to Accelerated Bridge CNI.name: the name of the network created.ipam(optional) : the configuration of the IP Address Management plugin. Required to designate an IP for a kernel interface.
Example configurations
The following examples show the config needed to set up basic Accelerated Bridge networking in a container. Each of the json config objects below can be placed in the .spec.config field of a Network Attachment Definition to integrate with Multus.
Basic config
This is the minimum configuration for Accelerated Bridge CNI. It applies an IP address using the host-local IPAM plugin in the range of the subnet provided.
{
"type": "accelerated-bridge",
"cniVersion": "0.3.1",
"name": "some-net",
"ipam": {
"type": "host-local",
"subnet": "10.56.217.0/24",
"routes": [{
"dst": "0.0.0.0/0"
}],
"gateway": "10.56.217.1"
}
}
Note: by default Accelerated Bridge CNI will use Linux Bridge with name cni0
Extended config
This configuration sets a number of extra parameters that may be key for Accelerated Bridge networks including a vlan tag, disabled spoof checking and enabled trust mode. These parameters are commonly set in more advanced Accelerated Bridge VF based networks.
{
"cniVersion": "0.3.1",
"name": "some-net-advanced",
"type": "accelerated-bridge",
"vlan": 1000,
"bridge": "br1",
"ipam": {
"type": "host-local",
"subnet": "10.56.217.0/24",
"routes": [{
"dst": "0.0.0.0/0"
}],
"gateway": "10.56.217.1"
}
}
Advanced Configuration
To learn more about available configuration parameters, check Accelerated Bridge CNI configuration reference guide
Contributing
To report a bug or request a feature, open an issue on this repo using one of the available templates.
Directories