简介

Kube vip为社区开源的一套Kubernetes HA解决方案，基于GO语言开发。轻量级，性能待验证。官方介绍如下：

Kubernetes Virtual IP and Load-Balancer for both control plane and Kubernetes services

The idea behind kube-vip is a small self-contained Highly-Available option for all environments, especially:

  • Bare-Metal
  • On-Prem
  • Edge (ARM / Raspberry PI)
  • Virtualisation
  • Pretty much anywhere else :)

Kubernetes 高可用

As an alternative to the more “traditional” approach of keepalived and haproxy, kube-vip implements both management of a virtual IP and load balancing in one service. Similar to option 2 above, kube-vip will be run as a static pod on the control plane nodes.

Like with keepalived, the hosts negotiating a virtual IP need to be in the same IP subnet. Similarly, like with haproxy, stream-based load-balancing allows TLS termination to be handled by the API Server instances behind it.

The configuration file /etc/kube-vip/config.yaml looks like this:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
localPeer:
  id: ${ID}
  address: ${IPADDR}
  port: 10000
remotePeers:
- id: ${PEER1_ID}
  address: ${PEER1_IPADDR}
  port: 10000
# [...]
vip: ${APISERVER_VIP}
gratuitousARP: true
singleNode: false
startAsLeader: ${IS_LEADER}
interface: ${INTERFACE}
loadBalancers:
- name: API Server Load Balancer
  type: tcp
  port: ${APISERVER_DEST_PORT}
  bindToVip: false
  backends:
  - port: ${APISERVER_SRC_PORT}
    address: ${HOST1_ADDRESS}
  # [...]

The bash style placeholders to expand are these:

  • ${ID} the current host’s symbolic name
  • ${IPADDR} the current host’s IP address
  • ${PEER1_ID} a symbolic name for the first vIP peer
  • ${PEER1_IPADDR} IP address for the first vIP peer
  • entries (id, address, port) for additional vIP peers can follow
  • ${APISERVER_VIP} is the virtual IP address negotiated between the kube-vip cluster hosts.
  • ${IS_LEADER} is true for exactly one leader and false for the rest
  • ${INTERFACE} is the network interface taking part in the negotiation of the virtual IP, e.g. eth0.
  • ${APISERVER_DEST_PORT} the port through which Kubernetes will talk to the API Server.
  • ${APISERVER_SRC_PORT} the port used by the API Server instances
  • ${HOST1_ADDRESS} the first load-balanced API Server host’s IP address
  • entries (port, address) for additional load-balanced API Server hosts can follow

To have the service started with the cluster, now the manifest kube-vip.yaml needs to be placed in /etc/kubernetes/manifests (create the directory first). It can be generated using the kube-vip docker image:

1
2
3
# docker run -it --rm plndr/kube-vip:0.1.1 /kube-vip sample manifest \
    | sed "s|plndr/kube-vip:'|plndr/kube-vip:0.1.1'|" \
    | sudo tee /etc/kubernetes/manifests/kube-vip.yaml

The result, /etc/kubernetes/manifests/kube-vip.yaml, will look like this:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  name: kube-vip
  namespace: kube-system
spec:
  containers:
  - command:
    - /kube-vip
    - start
    - -c
    - /vip.yaml
    image: 'plndr/kube-vip:0.1.1'
    name: kube-vip
    resources: {}
    securityContext:
      capabilities:
        add:
        - NET_ADMIN
        - SYS_TIME
    volumeMounts:
    - mountPath: /vip.yaml
      name: config
  hostNetwork: true
  volumes:
  - hostPath:
      path: /etc/kube-vip/config.yaml
    name: config
status: {}

With the services up, now the Kubernetes cluster can be bootstrapped using kubeadm init (see below).

Bootstrap the cluster

Now the actual cluster bootstrap as described in Creating Highly Available clusters with kubeadm can take place.

Note that, if ${APISERVER_DEST_PORT} has been configured to a value different from 6443 in the configuration above, kubeadm init needs to be told to use that port for the API Server. Assuming that in a new cluster port 8443 is used for the load-balanced API Server and a virtual IP with the DNS name vip.mycluster.local, an argument --control-plane-endpoint needs to be passed to kubeadm as follows:

1
# kubeadm init --control-plane-endpoint vip.mycluster.local:8443 [additional arguments ...]

参考文档