此处忽略k8s以及helm的安装。。。。
（1）查找neo4j安装包
helm search neo4j
NAME CHART VERSION APP VERSION DESCRIPTION
incubator/neo4j 0.3.0 3.2.3 Neo4j is the world's leading graph database
stable/neo4j 0.8.0 3.4.5 Neo4j is the world's leading graph database
（2）查看安装包配置项
helm inspect incubator/neo4j
可支持Causal cluster集群
（3）获取安装包
helm fetch stable/neo4j
（4）解压安装包
tar -xvzf neo4j-0.8.0.tgz && cd neo4j/
（5）创建storgeclass
vim nfs-client-class-neo4j.yaml
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: neo4j-nfs-storage
provisioner: fuseim.pri/ifs # or choose another name, must match deployment's env PROVISIONER_NAME'
fuseim.pri/ifs是provisioner，如何创建provisioner，自行查找，资料较多，使用storgeclass是一种动态创建PV以及PVC的方式。
创建：
kubectl create -f nfs-client-class-neo4j.yaml
storageclass.storage.k8s.io "neo4j-nfs-storage" created
查看：
kubectl get storageclass
NAME PROVISIONER AGE
neo4j-nfs-storage fuseim.pri/ifs 49s
（6）配置storageclass
vim values.yaml
core:
numberOfServers: 3
persistentVolume:
mountPath: /data
size: 20Gi
storageClass: "neo4j-nfs-storage"
storageClass指向刚才创建的storageclass的名称。
（7）配置neo4j参数
若需要修改neo4j的配置，通过以下步骤修改：
在templates/core-statefulset.yaml、templates/readreplicas-deployment.yaml通过在command:下添加类似于如下以NEO4J_为开头的内容来添加：
export NEO4J_dbms_shell_enabled=true
或者在containers下添加例如如下内容：
- name: NEO4J_dbms_shell_enabled
value: "true"
（8）安装neo4j
helm install ./neo4j --name neo4j （注意：这里不要指定neo4j的namespace，因为neo4j的charts包里将namespace默认写在在default下，若指定namespace，需要去neo4j/templates/core-statefulset.yaml等文件中将default改成neo4j）
NAME: neo4j
LAST DEPLOYED: Mon Jan 7 17:12:54 2019
NAMESPACE: default
STATUS: DEPLOYED
RESOURCES:
==> v1/Secret
NAME AGE
neo4j-neo4j-secrets 0s
==> v1/ConfigMap
neo4j-neo4j-tests 0s
==> v1/Service
neo4j-neo4j 0s
==> v1beta1/Deployment
neo4j-neo4j-replica 0s
==> v1beta1/StatefulSet
neo4j-neo4j-core 0s
==> v1/Pod(related)
NAME READY STATUS RESTARTS AGE
neo4j-neo4j-core-0 0/1 ContainerCreating 0 0s
NOTES:
We'll need to wait a few seconds for the Neo4j cluster to form.
We need to see this line in all of our pods' logs:
> Remote interface available at http://neo-helm-neo4j-core-2.neo-helm-neo4j.default.svc.cluster.local:7474/
We can see the content of the logs by running the following command:
kubectl logs -l "app=neo4j,component=core"
We can now run a query to find the topology of the cluster.
kubectl run -it --rm cypher-shell \
--image=neo4j:3.2.3-enterprise \
--restart=Never \
--namespace default \
--command -- ./bin/cypher-shell -u neo4j -p <password> --a neo4j-neo4j.default.svc.cluster.local "call dbms.cluster.overview()"
This will print out the addresses of the members of the cluster.
Note:
You'll need to substitute <password> with the password you set when installing the Helm package.
If you didn't set a password, one will be auto generated.
You can find the base64 encoded version of the password by running the following command:
kubectl get secrets neo4j-neo4j-secrets -o yaml
（9）验证
登录其中一台
kubectl exec -ti neo4j-neo4j-core-1 -- bash
./bin/cypher-shell
username: neo4j
password: ********** //密码请查看/var/lib/neo4j/conf/neo4j.conf文件的最后一行
neo4j> CREATE (n:Person {name: "dahua", age: 18});
neo4j> MATCH (a:Person) return a;
+------------------------------------+
| a |
+------------------------------------+
| (:Person {name: "dahua", age: 18}) |
+------------------------------------+
PS
下方是我个人订阅号，会一直更新各类技术文章，欢迎关注 ：）
