关于云计算:KubeSphere-集群配置-NFS-存储解决方案

作者：申红磊，QingCloud 容器解决方案架构师，开源我的项目爱好者，KubeSphere Member

在正式浏览本文之前，先情谊揭示一下：不建议您在生产环境中应用 NFS 存储（特地是 Kubernetes 1.20 或以上版本），起因如下：

• selfLink was empty 在 K8s 集群 v1.20 之前都存在，在 v1.20 之后被删除问题。
• 还有可能引起 failed to obtain lock 和 input/output error 等问题，从而导致 Pod CrashLoopBackOff。此外，局部利用不兼容 NFS，例如 Prometheus 等。

装置 NFS Server

# 装置 NFS 服务器端
$ sudo apt-get update #执行以下命令确保应用最新软件包
$ sudo apt-get install nfs-kernel-server
#装置 NFS 客户端
$ sudo apt-get install nfs-common
 
# yum
$ yum install -y nfs-utils

创立共享目录

先查看配置文件 /etc/exports：

$ cat /etc/exports
# /etc/exports: the access control list for filesystems which may be exported
#        to NFS clients.  See exports(5).
#
# Example for NFSv2 and NFSv3:
# /srv/homes       hostname1(rw,sync,no_subtree_check) hostname2(ro,sync,no_subtree_check)
#
# Example for NFSv4:
# /srv/nfs4        gss/krb5i(rw,sync,fsid=0,crossmnt,no_subtree_check)
# /srv/nfs4/homes  gss/krb5i(rw,sync,no_subtree_check)

创立共享指标并赋权：

# 目录 ksha3.2
$ mkdir /ksha3.2
$ chmod 777 /ksha3.2
#目录 demo
$ mkdir /demo
$ chmod 777 /demo
#目录 /home/ubuntu/nfs/ks3.1
$ mkdir /home/ubuntu/nfs/ks3.1
$ chmod 777 /home/ubuntu/nfs/ks3.1

增加到配置文件中 /etc/exports：

$ vi /etc/exports
....
/home/ubuntu/nfs/ks3.1 *(rw,sync,no_subtree_check)
/mount/ksha3.2 *(rw,sync,no_subtree_check)
/mount/demo *(rw,sync,no_subtree_check)
#/mnt/ks3.2 139.198.186.39(insecure,rw,sync,anonuid=500,anongid=500)
#/mnt/demo 139.198.167.103(rw,sync,no_subtree_check)
#此文件的配置格局为：< 输入目录 > [客户端 1 选项（拜访权限, 用户映射, 其余）] [客户端 2 选项（拜访权限, 用户映射, 其余）]

留神：如果共享目录创立有效或者脱漏创立，应用时会报异样如下：

$ mount.nfs: access denied by server while mounting 139.198.168.114:/mnt/demo1
#客户端挂载 nfs 共享目录的时候提醒 mount.nfs: access denied by server while mounting。#问题起因：服务器端的共享目录没有设置容许其他人拜访的权限，或者客户端挂载的目录没有权限。#解决办法：在服务器端批改共享目录的权限，胜利连贯。

验证

# 更新配置文件, 从新加载 /etc/exports 的配置：$ exportfs -rv
#在 nfs server 上测试 查看本机共享的目录：$ showmount -e 127.0.0.1
$ showmount -e localhost
 
$ showmount -e 127.0.0.1
Export list for 127.0.0.1:
/mount/demo            *
/mount/ksha3.2         *
/home/ubuntu/nfs/ks3.1 *

应用

在其它网络通的机器上，应用 NFS 共享目录。

装置客户端

如果不装置客户端会报：bad option; for several filesystems (e.g. nfs, cifs) you might need a /sbin/mount. helper program.

#apt-get
$ apt-get install nfs-common
#yum
$ yum install nfs-utils

在用户机器上进行测试：

# 本机创立挂载目录
#目录 ksha3.2
$ mkdir /root/aa
$ chmod 777 /root/aa
 
#应用 nfs client 上测试，将 /root/aa 目录挂载到远端 139.198.168.114 的共享存储目录 /mount/ksha3.2 上
$ mount -t nfs 139.198.168.114:/mount/ksha3.2 /root/aa

查看：

# 应用 df -h 查看
$ df -h | grep aa
139.198.168.114:/mount/ksha3.2  146G  5.3G  140G   4% /root/aa

解绑：

$ umount -t nfs 139.198.168.114:/mnt/ksha3.2 /root/aa
#如果卸载的时候提醒：umount:/mnt:device is busy；解决办法：须要退出挂载目录再进行卸载，或者是否 NFS server 宕机了
#须要强制卸载：mount –lf /mnt 
#此命令也能够：fuser –km /mnt 不倡议用 

KubeSphere 对接 NFS 动态分配器

可装置 NFS 客户端程序，为了不便用户对接 NFS 服务端，KubeSphere 可装置 NFS 动态分配器，反对动态分配存储卷，调配和回收存储卷过程简便，可对接一个或者多个 NFS 服务端。

当然也能够应用 Kubernetes 官网办法对接 NFS 服务端，这是一种动态调配存储卷办法，调配和回收存储卷过程简单，可对接多个 NFS 服务端。

前提条件

用户对接 NFS 服务端时应确保 KubeSphere 各节点有权限挂载 NFS 服务端文件夹。

操作步骤

以下步骤示例中，NFS 服务端 IP 为 139.198.168.114，NFS 共享文件夹为 /mnt/ksha3.2。

装置 NFS 动态分配器

首先请在这里下载 rbac.yaml，或者间接执行命令：

$ kubectl apply -f https://raw.githubusercontent.com/kubernetes-incubator/external-storage/master/nfs-client/deploy/rbac.yaml
serviceaccount/nfs-client-provisioner created
clusterrole.rbac.authorization.k8s.io/nfs-client-provisioner-runner created
clusterrolebinding.rbac.authorization.k8s.io/run-nfs-client-provisioner created
role.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created
rolebinding.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created

而后下载官网的 nfs provisoner 用处的 deployment。deployment 文件中有几处，请依据本人的状况做批改：

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nfs-client-provisioner
  labels:
    app: nfs-client-provisioner
  # replace with namespace where provisioner is deployed
  namespace: default
spec:
  replicas: 1
  strategy:
    type: Recreate
  selector:
    matchLabels:
      app: nfs-client-provisioner
  template:
    metadata:
      labels:
        app: nfs-client-provisioner
    spec:
      serviceAccountName: nfs-client-provisioner
      containers:
        - name: nfs-client-provisioner
          image: quay.io/external_storage/nfs-client-provisioner:latest
          volumeMounts:
            - name: nfs-client-root
              mountPath: /persistentvolumes
          env:
            - name: PROVISIONER_NAME
              value: nfs/provisioner-229
            - name: NFS_SERVER
              value: 139.198.168.114
            - name: NFS_PATH
              value: /mount/ksha3.2
      volumes:
        - name: nfs-client-root
          nfs:
            server: 139.198.168.114
            path: /mount/ksha3.2

执行创立：

$ kubectl apply -f deployment.yaml
deployment.apps/nfs-client-provisioner created

接下来下载官网的 class.yaml，而后创立 StorageClass，须要依据理论状况批改参数：

# 请依据上方 deployment 部署时候的 provisioner_name 做对应的批改，或者没有批改，就不必动
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: managed-nfs-storage
  annotations:
    "storageclass.kubernetes.io/is-default-class": "false"
provisioner: nfs/provisioner-229 # or choose another name, must match deployment's env PROVISIONER_NAME'
parameters:
  archiveOnDelete: "false"

执行创立：kubectl apply -f storageclass.yaml。

如果想让这个 NFS 作为默认的 Provisioner, 那么就增加如下的 annotation：

annotations:
  "storageclass.kubernetes.io/is-default-class": "true"

或者标记一个 StorageClass 为默认的 StorageClass, 你须要增加 / 设置注解 storageclass.kubernetes.io/is-default-class=true。

$ kubectl patch storageclass <your-class-name> -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'
 
# 查看集群中是否已存在 default Storage Class
$ kubectl get sc
NAME                         PROVISIONER                   AGE
glusterfs (default)         kubernetes.io/glusterfs        3d4h

验证装置后果

执行以下命令，查看 NFS 动态分配器容器组是否失常运行。

$ kubectl get po -A | grep nfs-client
default                        nfs-client-provisioner-7d69b9f45f-ks94m                           1/1     Running     0          9m3s

查看 NFS 存储类型

$ kubectl get sc managed-nfs-storage
NAME                  PROVISIONER           RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
managed-nfs-storage   nfs/provisioner-229   Delete          Immediate           false                  6m28s

创立和挂载 NFS 存储卷

当初能够通过动态创建 NFS 存储卷和工作负载挂载 NFS 存储卷了。

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: demo4nfs
  namespace: ddddd
  annotations:
    kubesphere.io/creator: admin
    volume.beta.kubernetes.io/storage-provisioner: nfs/provisioner-229
  finalizers:
    - kubernetes.io/pvc-protection
spec:
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 10Gi
  storageClassName: managed-nfs-storage
  volumeMode: Filesystem

unexpected error getting claim reference: selfLink was empty, can’t make reference

问题景象

应用 NFS 创立 PV 时，PVC 始终是处于 Pending 状态。

查看 PVC：

$ kubectl get pvc -n ddddd
NAME       STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS          AGE
demo4nfs   Bound    pvc-a561ce85-fc0d-42af-948e-6894ac000264   10Gi       RWO            managed-nfs-storage   32m

查看详细信息：

# 查看以后 pvc 的状态信息, 发现是在期待 volume 的创立
$ kubectl get pvc -n ddddd

查看 nfs-client-provisioner 的日志，是 seltlink was empty 的问题，selfLink was empty 在 K8s 集群 v1.20 之前都存在，在 v1.20 之后被删除，须要在 /etc/kubernetes/manifests/kube-apiserver.yaml 中增加参数。

$ kubectl get pod -n default
NAME                                      READY   STATUS    RESTARTS   AGE
nfs-client-provisioner-7d69b9f45f-ks94m   1/1     Running   0          27m
 
$ kubectl logs -f nfs-client-provisioner-7d69b9f45f-ks94m
I0622 09:41:33.606000       1 leaderelection.go:185] attempting to acquire leader lease  default/nfs-provisioner-229...
E0622 09:41:33.612745       1 event.go:259] Could not construct reference to: '&v1.Endpoints{TypeMeta:v1.TypeMeta{Kind:"", APIVersion:""}, ObjectMeta:v1.ObjectMeta{Name:"nfs-provisioner-229", GenerateName:"", Namespace:"default", SelfLink:"", UID:"e8f19e28-f17f-4b22-9bb8-d4cbe20c796b", ResourceVersion:"23803580", Generation:0, CreationTimestamp:v1.Time{Time:time.Time{wall:0x0, ext:63791487693, loc:(*time.Location)(0x1956800)}}, DeletionTimestamp:(*v1.Time)(nil), DeletionGracePeriodSeconds:(*int64)(nil), Labels:map[string]string(nil), Annotations:map[string]string{"control-plane.alpha.kubernetes.io/leader":"{\"holderIdentity\":\"nfs-client-provisioner-7d69b9f45f-ks94m_8081b417-f20f-11ec-bff3-f64a9f402fda\",\"leaseDurationSeconds\":15,\"acquireTime\":\"2022-06-22T09:41:33Z\",\"renewTime\":\"2022-06-22T09:41:33Z\",\"leaderTransitions\":0}"}, OwnerReferences:[]v1.OwnerReference(nil), Initializers:(*v1.Initializers)(nil), Finalizers:[]string(nil), ClusterName:""}, Subsets:[]v1.EndpointSubset(nil)}'due to:'selfLink was empty, can't make reference'. Will not report event: 'Normal' 'LeaderElection' 'nfs-client-provisioner-7d69b9f45f-ks94m_8081b417-f20f-11ec-bff3-f64a9f402fda became leader'
I0622 09:41:33.612829       1 leaderelection.go:194] successfully acquired lease default/nfs-provisioner-229
I0622 09:41:33.612973       1 controller.go:631] Starting provisioner controller nfs/provisioner-229_nfs-client-provisioner-7d69b9f45f-ks94m_8081b417-f20f-11ec-bff3-f64a9f402fda!
I0622 09:41:33.713170       1 controller.go:680] Started provisioner controller nfs/provisioner-229_nfs-client-provisioner-7d69b9f45f-ks94m_8081b417-f20f-11ec-bff3-f64a9f402fda!
I0622 09:53:33.461902       1 controller.go:987] provision "ddddd/demo4nfs" class "managed-nfs-storage": started
E0622 09:53:33.464213       1 controller.go:1004] provision "ddddd/demo4nfs" class "managed-nfs-storage": unexpected error getting claim reference: selfLink was empty, can't make reference
I0622 09:56:33.623717       1 controller.go:987] provision "ddddd/demo4nfs" class "managed-nfs-storage": started
E0622 09:56:33.625852       1 controller.go:1004] provision "ddddd/demo4nfs" class "managed-nfs-storage": unexpected error getting claim reference: selfLink was empty, can't make reference

解决办法

在 kube-apiserver.yaml 文件中增加参数 - --feature-gates=RemoveSelfLink=false。

# 应用命令查找一下 kube-apiserver.yaml 的地位
$ find / -name kube-apiserver.yaml
/data/kubernetes/manifests/kube-apiserver.yaml
#在文件中增加 - --feature-gates=RemoveSelfLink=false，如下图
$ cat /data/kubernetes/manifests/kube-apiserver.yaml
apiVersion: v1
kind: Pod
metadata:
  annotations:
    kubeadm.kubernetes.io/kube-apiserver.advertise-address.endpoint: 192.168.100.25:6443
  creationTimestamp: null
  labels:
    component: kube-apiserver
    tier: control-plane
  name: kube-apiserver
  namespace: kube-system
spec:
  containers:
  - command:
    - kube-apiserver
    - --feature-gates=RemoveSelfLink=false
    - --advertise-address=0.0.0.0
    - --allow-privileged=true
    - --authorization-mode=Node,RBAC
    - --client-ca-file=/etc/kubernetes/pki/ca.crt
    - --enable-admission-plugins=NodeRestriction
 
#增加之后应用 kubeadm 部署的集群会主动加载部署 pod
#kubeadm 装置的 apiserver 是 Static Pod，它的配置文件被批改后，立刻失效。#Kubelet 会监听该文件的变动，当您批改了 /etc/kubenetes/manifest/kube-apiserver.yaml 文件之后，kubelet 将主动终止原有的 #kube-apiserver-{nodename} 的 Pod，并主动创立一个应用了新配置参数的 Pod 作为代替。#如果您有多个 Kubernetes Master 节点，您须要在每一个 Master 节点上都批改该文件，并使各节点上的参数保持一致。#这里需注意如果 api-server 启动失败 需从新在执行一遍
$ kubectl apply -f /etc/kubernetes/manifests/kube-apiserver.yaml

GitHub 官网 ISSUES

• unexpected error getting claim reference: selfLink was empty, can’t make reference

  本文由博客一文多发平台 OpenWrite 公布！