通常情况下在k8s中，一个pod只有一个接口，用于集群网络中pod和pod通信，
而multus定义了一种crd(Kubernetes Network Custom Resource Definition)-NetworkAttachmentDefinition,可用来定义其他网络接口，使pod可以生成多个接口。
如下图，eth0为默认的集群网络中的接口，net0和net1是自定义的其他接口。
image.png
安装multus
安装multus前，k8s必须已经安装了一种cni用于连接pod集群网络，比如calico。
再安装multus，multus会使用之前的cni作为默认网络，将之前cni的配置放在/etc/cni/net.d/00-multus.conf。
再通过NetworkAttachmentDefinition定义其他的cni接口。
在pod的yaml的annotation中使用"k8s.v1.cni.cncf.io/networks" 引用，如果用逗号分开，同时添加多个接口。
参考官网，安装multus有两种方法，一种是手动安装(手动下载multus binary，手动创建sa，crd等)，另一种是通过daemonset自动安装。
下面采用自动安装方法
#下载源码
root@master:~/multus# git clone https://github.com/intel/multus-cni.git
#安装multus
root@master:~/multus# kubectl apply -f multus-cni-master/images/multus-daemonset.yml
customresourcedefinition.apiextensions.k8s.io/network-attachment-definitions.k8s.cni.cncf.io created
clusterrole.rbac.authorization.k8s.io/multus created
clusterrolebinding.rbac.authorization.k8s.io/multus created
serviceaccount/multus created
configmap/multus-cni-config created
daemonset.apps/kube-multus-ds-amd64 created
daemonset.apps/kube-multus-ds-ppc64le created
查看multus pod已经处于running状态
root@master:~/multus# kubectl get pod -n kube-system
NAME READY STATUS RESTARTS AGE
calico-kube-controllers-5b644bc49c-vrlmw 1/1 Running 0 7h45m
calico-node-5fhft 1/1 Running 0 7h45m
calico-node-8jpzr 1/1 Running 0 7h45m
calico-node-p8wxx 1/1 Running 0 7h45m
coredns-9d85f5447-4znmx 1/1 Running 4 42d
coredns-9d85f5447-fh667 1/1 Running 2 42d
etcd-master 1/1 Running 8 184d
kube-apiserver-master 1/1 Running 0 36h
kube-controller-manager-master 1/1 Running 8 184d
kube-multus-ds-amd64-469ls 1/1 Running 0 2s
kube-multus-ds-amd64-bzkts 1/1 Running 0 2s
kube-multus-ds-amd64-pj2p4 1/1 Running 0 2s
kube-proxy-l4wn7 1/1 Running 5 184d
kube-proxy-prhcm 1/1 Running 5 184d
kube-proxy-psxqt 1/1 Running 8 184d
kube-scheduler-master 1/1 Running 8 184d
network-attachment-definitions.k8s.cni.cncf.io 为multus创建的crd，用于定义其他网络接口
root@master:~/multus# kubectl describe CustomResourceDefinition network-attachment-definitions.k8s.cni.cncf.io
Name: network-attachment-definitions.k8s.cni.cncf.io
Namespace:
Labels: <none>
Annotations: kubectl.kubernetes.io/last-applied-configuration:
{"apiVersion":"apiextensions.k8s.io/v1","kind":"CustomResourceDefinition","metadata":{"annotations":{},"name":"network-attachment-definiti...
API Version: apiextensions.k8s.io/v1
Kind: CustomResourceDefinition
Metadata:
Creation Timestamp: 2020-08-16T21:28:15Z
Generation: 1
Resource Version: 12551106
Self Link: /apis/apiextensions.k8s.io/v1/customresourcedefinitions/network-attachment-definitions.k8s.cni.cncf.io
UID: 71df5215-ce14-41b3-bd6b-bf6bfa8198f5
Spec:
Conversion:
Strategy: None
Group: k8s.cni.cncf.io
Names:
Kind: NetworkAttachmentDefinition
List Kind: NetworkAttachmentDefinitionList
Plural: network-attachment-definitions
Short Names:
net-attach-def
Singular: network-attachment-definition
Scope: Namespaced
Versions:
Name: v1
Schema:
openAPIV3Schema:
Description: NetworkAttachmentDefinition is a CRD schema specified by the Network Plumbing Working Group to express the intent for attaching pods to one or more logical or physical networks. More information available at: https://github.com/k8snetworkplumbingwg/multi-net-spec
Properties:
Spec:
Description: NetworkAttachmentDefinition spec defines the desired state of a network attachment
Properties:
Config:
Description: NetworkAttachmentDefinition config is a JSON-formatted CNI configuration
Type: string
Type: object
Type: object
Served: true
Storage: true
Status:
Accepted Names:
Kind: NetworkAttachmentDefinition
List Kind: NetworkAttachmentDefinitionList
Plural: network-attachment-definitions
Short Names:
net-attach-def
Singular: network-attachment-definition
Conditions:
Last Transition Time: 2020-08-16T21:28:15Z
Message: no conflicts found
Reason: NoConflicts
Status: True
Type: NamesAccepted
Last Transition Time: 2020-08-16T21:28:15Z
Message: the initial names have been accepted
Reason: InitialNamesAccepted
Status: True
Type: Established
Stored Versions:
v1
Events: <none>
使用multus创建多个接口
首先创建一个NetworkAttachmentDefinition，指定macvlan类型
cat <<EOF | kubectl create -f -
apiVersion: "k8s.cni.cncf.io/v1"
kind: NetworkAttachmentDefinition
metadata:
name: macvlan-conf
spec:
config: '{
"cniVersion": "0.3.0",
"type": "macvlan",
"master": "ens3",
"mode": "bridge",
"ipam": {
"type": "host-local",
"subnet": "192.168.1.0/24",
"rangeStart": "192.168.1.200",
"rangeEnd": "192.168.1.216",
"routes": [
{ "dst": "0.0.0.0/0" }
],
"gateway": "192.168.1.1"
}
}'
EOF
root@master:~/multus# kubectl get net-attach-def
NAME AGE
macvlan-conf 25s
作为对比，先创建一个默认的pod
cat <<EOF | kubectl create -f -
apiVersion: v1
kind: Pod
metadata:
name: samplepod
spec:
containers:
- name: samplepod
command: ["/bin/ash", "-c", "trap : TERM INT; sleep infinity & wait"]
image: alpine
EOF
如下为pod默认的网络接口。
root@master:~/multus# kubectl exec -it samplepod -- ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 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
2: tunl0@NONE: <NOARP> mtu 1480 qdisc noop state DOWN qlen 1000
link/ipip 0.0.0.0 brd 0.0.0.0
4: eth0@if46: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1440 qdisc noqueue state UP
link/ether b2:95:21:e8:80:63 brd ff:ff:ff:ff:ff:ff
inet 10.24.166.141/32 scope global eth0
valid_lft forever preferred_lft forever
再使用macvlan-conf 创建一个pod
cat <<EOF | kubectl create -f -
apiVersion: v1
kind: Pod
metadata:
name: samplepod
annotations:
k8s.v1.cni.cncf.io/networks: macvlan-conf
spec:
containers:
- name: samplepod
command: ["/bin/ash", "-c", "trap : TERM INT; sleep infinity & wait"]
image: alpine
EOF
查看pod网络接口，多出来的net1为macvlan接口
root@master:~/multus# kubectl exec -it samplepod -- ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 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
2: tunl0@NONE: <NOARP> mtu 1480 qdisc noop state DOWN qlen 1000
link/ipip 0.0.0.0 brd 0.0.0.0
4: eth0@if43: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1440 qdisc noqueue state UP
link/ether 66:5c:86:29:da:d2 brd ff:ff:ff:ff:ff:ff
inet 10.24.166.139/32 scope global eth0
valid_lft forever preferred_lft forever
5: net1@tunl0: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP
link/ether ce:b5:43:8e:50:b2 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.200/24 scope global net1
valid_lft forever preferred_lft forever
可以指定多次macvlan-conf来创建更多接口
cat <<EOF | kubectl create -f -
apiVersion: v1
kind: Pod
metadata:
name: samplepod
annotations:
k8s.v1.cni.cncf.io/networks: macvlan-conf,macvlan-conf
spec:
containers:
- name: samplepod
command: ["/bin/ash", "-c", "trap : TERM INT; sleep infinity & wait"]
image: alpine
EOF
多出来的net1和net2即为macvlan接口
root@master:~/multus# kubectl exec -it samplepod -- ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 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
2: tunl0@NONE: <NOARP> mtu 1480 qdisc noop state DOWN qlen 1000
link/ipip 0.0.0.0 brd 0.0.0.0
4: eth0@if45: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1440 qdisc noqueue state UP
link/ether 52:f8:ed:35:c0:77 brd ff:ff:ff:ff:ff:ff
inet 10.24.166.140/32 scope global eth0
valid_lft forever preferred_lft forever
5: net1@tunl0: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP
link/ether 6a:1a:1f:0f:0d:6e brd ff:ff:ff:ff:ff:ff
inet 192.168.1.201/24 scope global net1
valid_lft forever preferred_lft forever
6: net2@tunl0: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP
link/ether 2a:6b:f3:08:d3:d2 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.202/24 scope global net2
valid_lft forever preferred_lft forever
参考
如下链接都来自multus官方文档，写的很详细
https://github.com/intel/multus-cni/blob/master/doc/how-to-use.md
https://github.com/intel/multus-cni/blob/master/doc/quickstart.md
https://github.com/intel/multus-cni/blob/master/doc/configuration.md
网友评论