高可用多节点安装。
openebs、nfs-client pod 启动失败。
2 个 pod 均报错：
provisioner.go:247] Error getting server version: Get "https://10.233.0.1:443/version?timeout=32s": dial tcp 10.233.0.1:443: i/o timeout
但在任意节点上均可获取到 version 信息：
curl -k https://10.233.0.1:443/version?timeout=32s
{
"major": "1",
"minor": "20",
"gitVersion": "v1.20.6",
"gitCommit": "8a62859e515889f07e3e3be6a1080413f17cf2c3",
"gitTreeState": "clean",
"buildDate": "2021-04-15T03:19:55Z",
"goVersion": "go1.15.10",
"compiler": "gc",
"platform": "linux/amd64"
}
详细信息：
7 个 VirtualBox 虚拟机
虚机配置：4C/8G
操作系统：Debian 10.10
虚机网络：
- 桥接网络：172.31.31.0/24
- 虚机内部网络：192.168.168.0/24
172.31.31.70/192.168.168.71：负载均衡(Nginx TCP proxy)、NFS Server
172.31.31.71/192.168.168.72：Master 1
172.31.31.72/192.168.168.73：Master 2
172.31.31.73/192.168.168.74：Master 3
172.31.31.74/192.168.168.75：Node 1
172.31.31.75/192.168.168.76：Node 2
172.31.31.76/192.168.168.70：Node 3
安装方式：
高可用多节点安装
在线安装(KKZONE=cn)
kubekey 全套安装 K8S
kubekey 版本：v1.1.1
kubesphere 版本：v3.1.1
K8S 版本：v1.20.6
config-sample.yaml 中启用了 nfs-client addon，但并未设置 nfs-client 为默认 sc (defaultClass: false)
所以安装过程中还会自动安装 openebs。
经测试 nfs-client 设置为 defaultClass: true（不安装 openebs）或者，不启用 nfs-client，只安装默认的 openebs 都会引起上述报错。
以下是主要配置文件：
nfs-client.yaml
nfs:
server: "192.168.168.70" # This is the server IP address. Replace it with your own.
path: "/data/nfs" # Replace the exported directory with your own.
storageClass:
defaultClass: false
config-sample.yaml
apiVersion: kubekey.kubesphere.io/v1alpha1
kind: Cluster
metadata:
name: sample
spec:
hosts:
- {name: master1, address: 172.31.31.71, internalAddress: 192.168.168.71, user: root, password: xxx}
- {name: master2, address: 172.31.31.72, internalAddress: 192.168.168.72, user: root, password: xxx}
- {name: master3, address: 172.31.31.73, internalAddress: 192.168.168.73, user: root, password: xxx}
- {name: node1, address: 172.31.31.74, internalAddress: 192.168.168.74, user: root, password: xxx}
- {name: node2, address: 172.31.31.75, internalAddress: 192.168.168.75, user: root, password: xxx}
- {name: node3, address: 172.31.31.76, internalAddress: 192.168.168.76, user: root, password: xxx}
roleGroups:
etcd:
- master1
- master2
- master3
master:
- master1
- master2
- master3
worker:
- node1
- node2
- node3
controlPlaneEndpoint:
domain: lb.kubesphere.local
address: "172.31.31.70"
port: 6443
kubernetes:
version: v1.20.6
imageRepo: kubesphere
clusterName: cluster.local
network:
plugin: calico
kubePodsCIDR: 10.233.64.0/18
kubeServiceCIDR: 10.233.0.0/18
registry:
registryMirrors: []
insecureRegistries: []
addons:
- name: nfs-client
namespace: kube-system
sources:
chart:
name: nfs-client-provisioner
repo: https://charts.kubesphere.io/main
valuesFile: /usr/local/kubekey/nfs-client.yaml
kubekey 安装过程报错信息：
[master1 172.31.31.71] MSG:
namespace/kubesphere-system unchanged
serviceaccount/ks-installer unchanged
customresourcedefinition.apiextensions.k8s.io/clusterconfigurations.installer.kubesphere.io unchanged
clusterrole.rbac.authorization.k8s.io/ks-installer unchanged
clusterrolebinding.rbac.authorization.k8s.io/ks-installer unchanged
deployment.apps/ks-installer unchanged
clusterconfiguration.installer.kubesphere.io/ks-installer created
WARN[16:14:05 CST] Task failed ...
WARN[16:14:05 CST] error: KubeSphere startup timeout.
Error: Failed to deploy kubesphere: KubeSphere startup timeout.
kubectl get pod –all-namespaces
以下 2 个 pod 一直 Error
nfs-client-nfs-client-provisioner-794ffc57-qm6kc
openebs-localpv-provisioner-5cddd6cbfc-lbctr
报错信息：
1 provisioner.go:247] Error getting server version: Get "https://10.233.0.1:443/version?timeout=32s": dial tcp 10.233.0.1:443: i/o timeout
