k8s部署redis cluster集羣

k8s以StatefulSet方式部署redis cluster集羣：

1. 部署nfs

2. 創建pv

3. 部署redis

4. 初始化redis集羣

---

• 主機說明：

系統	ip	角色	cpu	內存	hostname
CentOS 7.8	192.168.30.128	master	>=2	>=2G	master1
CentOS 7.8	192.168.30.129	master	>=2	>=2G	master2
CentOS 7.8	192.168.30.130	node	>=2	>=2G	node1
CentOS 7.8	192.168.30.131	node	>=2	>=2G	node2
CentOS 7.8	192.168.30.132	node	>=2	>=2G	node3

kubectl get node

NAME      STATUS   ROLES    AGE   VERSION
master1   Ready    master   33d   v1.14.0
master2   Ready    master   33d   v1.14.0
node1     Ready    <none>   33d   v1.14.0
node2     Ready    <none>   33d   v1.14.0
node3     Ready    <none>   33d   v1.14.0

---

部署nfs

在master2節點上做nfs共享，

yum -y install nfs-utils rpcbind

mkdir -p /data/redis/{cluster0,cluster1,cluster2,cluster3,cluster4,cluster5}

vim /etc/exports

/data/redis/cluster0 192.168.30.0/24(rw,sync,no_root_squash)
/data/redis/cluster1 192.168.30.0/24(rw,sync,no_root_squash)
/data/redis/cluster2 192.168.30.0/24(rw,sync,no_root_squash)
/data/redis/cluster3 192.168.30.0/24(rw,sync,no_root_squash)
/data/redis/cluster4 192.168.30.0/24(rw,sync,no_root_squash)
/data/redis/cluster5 192.168.30.0/24(rw,sync,no_root_squash)

chmod -R 755 /data/redis

exportfs -arv

systemctl enable rpcbind && systemctl start rpcbind

systemctl enable nfs && systemctl start nfs

nfs部署完畢。對於需要使用nfs的node節點，都要安裝nfs：

yum -y install nfs-utils

---

創建pv

• 創建pv：

mkdir -p /home/k8s/redis  && cd /home/k8s/redis

vim pv.yaml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv0
spec:
  capacity:
    storage: 1Gi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.30.129
    path: /data/redis/cluster0

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv1
spec:
  capacity:
    storage: 1Gi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.30.129
    path: /data/redis/cluster1

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv2
spec:
  capacity:
    storage: 1Gi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.30.129
    path: /data/redis/cluster2

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv3
spec:
  capacity:
    storage: 1Gi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.30.129
    path: /data/redis/cluster3

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv4
spec:
  capacity:
    storage: 1Gi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.30.129
    path: /data/redis/cluster4

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv5
spec:
  capacity:
    storage: 1Gi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.30.129
    path: /data/redis/cluster5

kubectl apply -f pv.yaml

kubectl get pv

NAME      CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM   STORAGECLASS   REASON   AGE
nfs-pv0   1Gi        RWX            Retain           Available                                   3s
nfs-pv1   1Gi        RWX            Retain           Available                                   3s
nfs-pv2   1Gi        RWX            Retain           Available                                   3s
nfs-pv3   1Gi        RWX            Retain           Available                                   3s
nfs-pv4   1Gi        RWX            Retain           Available                                   3s
nfs-pv5   1Gi        RWX            Retain           Available                                   3s

pv創建完畢。

---

部署redis

vim namespace.yaml

apiVersion: v1
kind: Namespace
metadata:
  name: public-service

vim redis.conf

appendonly yes
cluster-enabled yes
cluster-config-file /var/lib/redis/nodes.conf
cluster-node-timeout 5000
dir /var/lib/redis
port 6379

vim redis.yaml

apiVersion: v1
kind: Service
metadata:
  name: redis
  namespace: public-service
  labels:
    app: redis
spec:
  selector:
    app: redis
    appCluster: redis-cluster
  ports:
  - name: redis
    port: 6379
  clusterIP: None
  
---
apiVersion: v1
kind: Service
metadata:
  name: redis-access
  namespace: public-service
  labels:
    app: redis
spec:
  selector:
    app: redis
    appCluster: redis-cluster
  ports:
  - name: redis-access
    protocol: TCP
    port: 6379
    targetPort: 6379

---
apiVersion: apps/v1beta1
kind: StatefulSet
metadata:
  name: redis
  namespace: public-service
spec:
  serviceName: redis
  replicas: 6
  template:
    metadata:
      labels:
        app: redis
        appCluster: redis-cluster
    spec:
      terminationGracePeriodSeconds: 20
      affinity:
        podAntiAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
          - weight: 100
            podAffinityTerm:
              labelSelector:
                matchExpressions:
                - key: app
                  operator: In
                  values:
                  - redis
              topologyKey: kubernetes.io/hostname
      containers:
      - name: redis
        image: redis:latest
        command:
          - "redis-server"
        args:
          - "/etc/redis/redis.conf"
          - "--protected-mode"
          - "no"
        resources:
          requests:
            cpu: "500m"
            memory: "500Mi"
        ports:
            - containerPort: 6379
              name: redis
              protocol: TCP
            - containerPort: 16379
              name: cluster
              protocol: TCP
        volumeMounts:
          - name: conf
            mountPath: /etc/redis
          - name: data
            mountPath: /var/lib/redis
      volumes:
      - name: conf
        configMap:
          name: redis-conf
          items:
            - key: redis.conf
              path: redis.conf
  volumeClaimTemplates:
  - metadata:
      name: data
      namespace: public-service
    spec:
      accessModes: [ "ReadWriteMany" ]
      resources:
        requests:
          storage: 1Gi

kubectl apply -f namespace.yaml

kubectl create configmap redis-conf --from-file=redis.conf -n public-service

kubectl apply -f redis.yaml

kubectl get svc -n public-service

NAME           TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)    AGE
redis          ClusterIP   None          <none>        6379/TCP   20s
redis-access   ClusterIP   10.96.2.100   <none>        6379/TCP   20s

kubectl get pod -n public-service

NAME      READY   STATUS    RESTARTS   AGE
redis-0   1/1     Running   0          2m43s
redis-1   1/1     Running   0          2m18s
redis-2   1/1     Running   0          108s
redis-3   1/1     Running   0          80s
redis-4   1/1     Running   0          48s
redis-5   1/1     Running   0          30s

kubectl get pvc -n public-service

NAME           STATUS   VOLUME    CAPACITY   ACCESS MODES   STORAGECLASS   AGE
data-redis-0   Bound    nfs-pv4   1Gi        RWX                           3m4s
data-redis-1   Bound    nfs-pv0   1Gi        RWX                           2m39s
data-redis-2   Bound    nfs-pv1   1Gi        RWX                           2m9s
data-redis-3   Bound    nfs-pv2   1Gi        RWX                           101s
data-redis-4   Bound    nfs-pv3   1Gi        RWX                           69s
data-redis-5   Bound    nfs-pv5   1Gi        RWX                           51s

redis-access這個service方便集羣內訪問redis集羣，redis部署完畢。

---

初始化redis集羣

使用Redis-tribe工具進行集羣的初始化。

• 下載redis-tribe：

kubectl run -it ubuntu --image=ubuntu --restart=Never -n public-service bash

root@ubuntu:/# cat > /etc/apt/sources.list << EOF
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial main restricted
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial-updates main restricted
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial universe
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial-updates universe
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial multiverse
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial-updates multiverse
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial-backports main restricted universe multiverse
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial-security main restricted
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial-security universe
deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ xenial-security multiverse
EOF

root@ubuntu:/# apt-get update

root@ubuntu:/# apt-get install -y libncursesw5 libreadline6 libtinfo5 --allow-remove-essential

root@ubuntu:/# apt-get install -y libpython2.7-stdlib python2.7 python-pip redis-tools dnsutils

root@ubuntu:/# pip install --upgrade pip

root@ubuntu:/# pip install redis-trib==0.5.1

• 初始化集羣：

root@ubuntu:/# redis-trib.py create \
  `dig +short redis-0.redis.public-service.svc.cluster.local`:6379 \
  `dig +short redis-1.redis.public-service.svc.cluster.local`:6379 \
  `dig +short redis-2.redis.public-service.svc.cluster.local`:6379
  
root@ubuntu:/# redis-trib.py replicate \
  --master-addr `dig +short redis-0.redis.public-service.svc.cluster.local`:6379 \
  --slave-addr `dig +short redis-3.redis.public-service.svc.cluster.local`:6379

root@ubuntu:/# redis-trib.py replicate \
  --master-addr `dig +short redis-1.redis.public-service.svc.cluster.local`:6379 \
  --slave-addr `dig +short redis-4.redis.public-service.svc.cluster.local`:6379

root@ubuntu:/# redis-trib.py replicate \
  --master-addr `dig +short redis-2.redis.public-service.svc.cluster.local`:6379 \
  --slave-addr `dig +short redis-5.redis.public-service.svc.cluster.local`:6379
  
root@ubuntu:/# exit

• 查看集羣：

kubectl exec -it -n public-service redis-0 bash

root@redis-0:/data# redis-cli -c

127.0.0.1:6379> CLUSTER NODES               #列出節點信息

aac2b3d320da67eedf3512ed0e38a1cdce5bc8fe 172.10.2.55:6379@16379 slave 7c4d60cf32685484ea6c5cb4493a937dfbf6b8a5 0 1592276224727 3 connected
2efad514b2f3c7fe4530dd6dc63c0df8ffdb793d 172.10.2.54:6379@16379 master - 0 1592276224224 1 connected 0-5461
524f03526a4b683d7d4de19296431810bfdc22cf 172.10.3.60:6379@16379 slave df5bc3c2e2851d63cdb9f762efde6e1b0d38efed 0 1592276223117 5 connected
7c4d60cf32685484ea6c5cb4493a937dfbf6b8a5 172.10.4.77:6379@16379 myself,master - 0 1592276224000 2 connected 5462-10922
df5bc3c2e2851d63cdb9f762efde6e1b0d38efed 172.10.3.59:6379@16379 master - 0 1592276223217 0 connected 10923-16383
c1dbaaef4a583e372c43eed52c22cd9ad7184d18 172.10.4.78:6379@16379 slave 2efad514b2f3c7fe4530dd6dc63c0df8ffdb793d 0 1592276223719 4 connected

127.0.0.1:6379> CLUSTER INFO                #集羣狀態

cluster_state:ok
cluster_slots_assigned:16384
cluster_slots_ok:16384
cluster_slots_pfail:0
cluster_slots_fail:0
cluster_known_nodes:6
cluster_size:3
cluster_current_epoch:5
cluster_my_epoch:2
cluster_stats_messages_ping_sent:514
cluster_stats_messages_pong_sent:491
cluster_stats_messages_meet_sent:2
cluster_stats_messages_sent:1007
cluster_stats_messages_ping_received:491
cluster_stats_messages_pong_received:516
cluster_stats_messages_received:1007

redis集羣初始化完成，已經形成3主3從的cluster集羣。

• 寫入數據：

127.0.0.1:6379> SET key1 aaa
OK

127.0.0.1:6379> SET key2 bbb
-> Redirected to slot [4998] located at 172.10.2.54:6379
OK

172.10.2.54:6379> SET key3 ccc
OK

kubectl exec -it -n public-service redis-4 bash

root@redis-4:/data# redis-cli -c

127.0.0.1:6379> GET key1
-> Redirected to slot [9189] located at 172.10.4.77:6379
"aaa"

172.10.4.77:6379> GET key2
-> Redirected to slot [4998] located at 172.10.2.54:6379
"bbb"

172.10.2.54:6379> GET key3
"ccc"

可以看出redis cluster集羣是去中心化的，每個節點都是平等的，連接哪個節點都可以獲取和設置數據。

• 主從切換：

選擇redis-2模擬宕掉，測試主從切換，

kubectl get pod -n public-service -o wide

NAME      READY   STATUS      RESTARTS   AGE   IP            NODE    NOMINATED NODE   READINESS GATES
redis-0   1/1     Running     0          62m   172.10.4.77   node1   <none>           <none>
redis-1   1/1     Running     0          62m   172.10.2.54   node2   <none>           <none>
redis-2   1/1     Running     0          61m   172.10.3.59   node3   <none>           <none>
redis-3   1/1     Running     0          61m   172.10.2.55   node2   <none>           <none>
redis-4   1/1     Running     0          61m   172.10.4.78   node1   <none>           <none>
redis-5   1/1     Running     0          61m   172.10.3.60   node3   <none>           <none>
ubuntu    0/1     Completed   0          57m   172.10.2.56   node2   <none>           <none>

kubectl exec -it -n public-service redis-2 bash

root@redis-2:/data# redis-cli -c

127.0.0.1:6379> ROLE

1) "master"
2) (integer) 2898
3) 1) 1) "172.10.3.60"
      2) "6379"
      3) "2898"

可以看到，redis-2是master，它的slave是172.10.3.60，即redis-5。

kubectl delete pod -n public-service redis-2                #模擬節點宕掉

kubectl get pod -n public-service redis-2 -o wide

NAME      READY   STATUS    RESTARTS   AGE   IP            NODE    NOMINATED NODE   READINESS GATES
redis-2   1/1     Running   0          38s   172.10.3.61   node3   <none>           <none>

kubectl exec -it -n public-service redis-2 bash

root@redis-2:/data# redis-cli -c

127.0.0.1:6379> ROLE

1) "slave"
2) "172.10.3.60"
3) (integer) 6379
4) "connected"
5) (integer) 3430

kubectl exec -it -n public-service redis-5 bash

root@redis-5:/data# redis-cli -c

127.0.0.1:6379> ROLE

1) "master"
2) (integer) 3584
3) 1) 1) "172.10.3.61"
      2) "6379"
      3) "3570"
      
127.0.0.1:6379> CLUSTER NODES

aac2b3d320da67eedf3512ed0e38a1cdce5bc8fe 172.10.2.55:6379@16379 slave 7c4d60cf32685484ea6c5cb4493a937dfbf6b8a5 0 1592278859530 2 connected
2efad514b2f3c7fe4530dd6dc63c0df8ffdb793d 172.10.2.54:6379@16379 master - 0 1592278859000 1 connected 0-5461
c1dbaaef4a583e372c43eed52c22cd9ad7184d18 172.10.4.78:6379@16379 slave 2efad514b2f3c7fe4530dd6dc63c0df8ffdb793d 0 1592278859000 1 connected
524f03526a4b683d7d4de19296431810bfdc22cf 172.10.3.60:6379@16379 myself,master - 0 1592278857000 6 connected 10923-16383
7c4d60cf32685484ea6c5cb4493a937dfbf6b8a5 172.10.4.77:6379@16379 master - 0 1592278858021 2 connected 5462-10922
df5bc3c2e2851d63cdb9f762efde6e1b0d38efed 172.10.3.61:6379@16379 slave 524f03526a4b683d7d4de19296431810bfdc22cf 0 1592278859000 6 connected

可以看到，redis-2在重啓之後變爲slave，而它之前的slave——redis-5變爲master，而且是新redis-2的master。

集羣的主從切換沒有問題。k8s部署redis cluster集羣完成。

---