摘要：ES集群是进行大数据存储和分析，快速检索的利器，本文简述了ES的集群架构，并提供了在Kubernetes中快速部署ES集群的样例；对ES集群的监控运维工具进行了介绍，并提供了部分问题定位经验，最后总结了常用ES集群的API调用方法。
本文分享自华为云社区《Kubernetes中部署ES集群及运维》，原文作者：minucas。
ES集群架构：
ES集群分为单点模式和集群模式，其中单点模式一般在生产环境不推荐使用，推荐使用集群模式部署。其中集群模式又分为Master节点与Data节点由同一个节点承担，以及Master节点与Data节点由不同节点承担的部署模式。Master节点与Data节点分开的部署方式可靠性更强。下图为ES集群的部署架构图：
采用K8s进行ES集群部署：
1、采用k8s statefulset部署，可快速的进行扩缩容es节点，本例子采用 3 Master Node + 12 Data Node 方式部署
2、通过k8s service配置了对应的域名和服务发现，确保集群能自动联通和监控
kubectl -s http://ip:port create -f es-master.yaml kubectl -s http://ip:port create -f es-data.yaml kubectl -s http://ip:port create -f es-service.yaml
es-master.yaml:
apiVersion: apps/v1 kind: StatefulSet metadata: labels: addonmanager.kubernetes.io/mode: Reconcile k8s-app: es kubernetes.io/cluster-service: "true" version: v6.2.5 name: es-master namespace: default spec: podManagementPolicy: OrderedReady replicas: 3 revisionHistoryLimit: 10 selector: matchLabels: k8s-app: es version: v6.2.5 serviceName: es template: metadata: labels: k8s-app: camp-es kubernetes.io/cluster-service: "true" version: v6.2.5 spec: containers: - env: - name: NAMESPACE valueFrom: fieldRef: apiVersion: v1 fieldPath: metadata.namespace - name: ELASTICSEARCH_SERVICE_NAME value: es - name: NODE_MASTER value: "true" - name: NODE_DATA value: "false" - name: ES_HEAP_SIZE value: 4g - name: ES_JAVA_OPTS value: -Xmx4g -Xms4g - name: cluster.name value: es image: elasticsearch:v6.2.5 imagePullPolicy: Always name: es ports: - containerPort: 9200 hostPort: 9200 name: db protocol: TCP - containerPort: 9300 hostPort: 9300 name: transport protocol: TCP resources: limits: cpu: "6" memory: 12Gi requests: cpu: "4" memory: 8Gi securityContext: capabilities: add: - IPC_LOCK - SYS_RESOURCE volumeMounts: - mountPath: /data name: es - command: - /bin/elasticsearch_exporter - -es.uri=http://localhost:9200 - -es.all=true image: elasticsearch_exporter:1.0.2 imagePullPolicy: IfNotPresent livenessProbe: failureThreshold: 3 httpGet: path: /health port: 9108 scheme: HTTP initialDelaySeconds: 30 periodSeconds: 10 successThreshold: 1 timeoutSeconds: 10 name: es-exporter ports: - containerPort: 9108 hostPort: 9108 protocol: TCP readinessProbe: failureThreshold: 3 httpGet: path: /health port: 9108 scheme: HTTP initialDelaySeconds: 10 periodSeconds: 10 successThreshold: 1 timeoutSeconds: 10 resources: limits: cpu: 100m memory: 128Mi requests: cpu: 25m memory: 64Mi securityContext: capabilities: drop: - SETPCAP - MKNOD - AUDIT_WRITE - CHOWN - NET_RAW - DAC_OVERRIDE - FOWNER - FSETID - KILL - SETGID - SETUID - NET_BIND_SERVICE - SYS_CHROOT - SETFCAP readOnlyRootFilesystem: true dnsPolicy: ClusterFirst initContainers: - command: - /sbin/sysctl - -w - vm.max_map_count=262144 image: alpine:3.6 imagePullPolicy: IfNotPresent name: elasticsearch-logging-init resources: {} securityContext: privileged: true restartPolicy: Always schedulerName: default-scheduler securityContext: {} volumes: - hostPath: path: /Data/es type: DirectoryOrCreate name: es
es-data.yaml
apiVersion: apps/v1 kind: StatefulSet metadata: labels: addonmanager.kubernetes.io/mode: Reconcile k8s-app: es kubernetes.io/cluster-service: "true" version: v6.2.5 name: es-data namespace: default spec: podManagementPolicy: OrderedReady replicas: 12 revisionHistoryLimit: 10 selector: matchLabels: k8s-app: es version: v6.2.5 serviceName: es template: metadata: labels: k8s-app: es kubernetes.io/cluster-service: "true" version: v6.2.5 spec: containers: - env: - name: NAMESPACE valueFrom: fieldRef: apiVersion: v1 fieldPath: metadata.namespace - name: ELASTICSEARCH_SERVICE_NAME value: es - name: NODE_MASTER value: "false" - name: NODE_DATA value: "true" - name: ES_HEAP_SIZE value: 16g - name: ES_JAVA_OPTS value: -Xmx16g -Xms16g - name: cluster.name value: es image: elasticsearch:v6.2.5 imagePullPolicy: Always name: es ports: - containerPort: 9200 hostPort: 9200 name: db protocol: TCP - containerPort: 9300 hostPort: 9300 name: transport protocol: TCP resources: limits: cpu: "8" memory: 32Gi requests: cpu: "7" memory: 30Gi securityContext: capabilities: add: - IPC_LOCK - SYS_RESOURCE volumeMounts: - mountPath: /data name: es - command: - /bin/elasticsearch_exporter - -es.uri=http://localhost:9200 - -es.all=true image: elasticsearch_exporter:1.0.2 imagePullPolicy: IfNotPresent livenessProbe: failureThreshold: 3 httpGet: path: /health port: 9108 scheme: HTTP initialDelaySeconds: 30 periodSeconds: 10 successThreshold: 1 timeoutSeconds: 10 name: es-exporter ports: - containerPort: 9108 hostPort: 9108 protocol: TCP readinessProbe: failureThreshold: 3 httpGet: path: /health port: 9108 scheme: HTTP initialDelaySeconds: 10 periodSeconds: 10 successThreshold: 1 timeoutSeconds: 10 resources: limits: cpu: 100m memory: 128Mi requests: cpu: 25m memory: 64Mi securityContext: capabilities: drop: - SETPCAP - MKNOD - AUDIT_WRITE - CHOWN - NET_RAW - DAC_OVERRIDE - FOWNER - FSETID - KILL - SETGID - SETUID - NET_BIND_SERVICE - SYS_CHROOT - SETFCAP readOnlyRootFilesystem: true dnsPolicy: ClusterFirst initContainers: - command: - /sbin/sysctl - -w - vm.max_map_count=262144 image: alpine:3.6 imagePullPolicy: IfNotPresent name: elasticsearch-logging-init resources: {} securityContext: privileged: true restartPolicy: Always schedulerName: default-scheduler securityContext: {} volumes: - hostPath: path: /Data/es type: DirectoryOrCreate name: es
es-service.yaml
apiVersion: v1 kind: Service metadata: labels: addonmanager.kubernetes.io/mode: Reconcile k8s-app: es kubernetes.io/cluster-service: "true" kubernetes.io/name: Elasticsearch name: es namespace: default spec: clusterIP: None ports: - name: es port: 9200 protocol: TCP targetPort: 9200 - name: exporter port: 9108 protocol: TCP targetPort: 9108 selector: k8s-app: es sessionAffinity: None type: ClusterIP
ES集群监控
工欲善其事必先利其器，中间件的运维首先要有充分的监控手段，ES集群的监控常用的三种监控手段：exporter、eshead、kopf，由于ES集群是采用k8s架构部署，很多特性都会结合k8s来开展
Grafana监控
通过k8s部署es-exporter将监控metrics导出，prometheus采集监控数据，grafana定制dashboard展示
ES-head组件
github地址：https://github.com/mobz/elasticsearch-head
ES-head组件可通过谷歌浏览器应用商店搜索安装，使用Chrome插件可查看ES集群的情况
Cerebro（kopf）组件
github地址：https://github.com/lmenezes/cerebro
ES集群问题处理
ES配置
资源配置：关注ES的CPU、Memory以及Heap Size，Xms Xmx的配置，建议如机器是8u32g内存的情况下，堆内存和Xms Xmx配置为50%，官网建议单个node的内存不要超过64G
索引配置：由于ES检索通过索引来定位，检索的时候ES会将相关的索引数据装载到内存中加快检索速度，因此合理的对索引进行设置对ES的性能影响很大，当前我们通过按日期创建索引的方法（个别数据量小的可不分割索引）
ES负载
CPU和Load比较高的节点重点关注，可能的原因是shard分配不均匀，此时可手动讲不均衡的shard relocate一下
shard配置
shard配置最好是data node数量的整数倍，shard数量不是越多越好，应该按照索引的数据量合理进行分片，确保每个shard不要超过单个data node分配的堆内存大小，比如数据量最大的index单日150G左右，分为24个shard，计算下来单个shard大小大概6-7G左右
副本数建议为1，副本数过大，容易导致数据的频繁relocate，加大集群负载
删除异常index
curl -X DELETE "10.64.xxx.xx:9200/szv-prod-ingress-nginx-2021.05.01"
索引名可使用进行正则匹配进行批量删除，如：-2021.05.*
节点负载高的另一个原因
在定位问题的时候发现节点数据shard已经移走但是节点负载一直下不去，登入节点使用top命令发现节点kubelet的cpu占用非常高，重启kubelet也无效，重启节点后负载才得到缓解
ES集群常规运维经验总结（参考官网）
查看集群健康状态
ES集群的健康状态分为三种：Green、Yellow、Red。
- Green(绿色)：集群健康；
- Yellow(黄色)：集群非健康，但在负载允许范围内可自动rebalance恢复；
- Red(红色)：集群存在问题，有部分数据未就绪，至少有一个主分片未分配成功。
可通过API查询集群的健康状态及未分配的分片：
GET _cluster/health { "cluster_name": "camp-es", "status": "green", "timed_out": false, "number_of_nodes": 15, "number_of_data_nodes": 12, "active_primary_shards": 2176, "active_shards": 4347, "relocating_shards": 0, "initializing_shards": 0, "unassigned_shards": 0, "delayed_unassigned_shards": 0, "number_of_pending_tasks": 0, "number_of_in_flight_fetch": 0, "task_max_waiting_in_queue_millis": 0, "active_shards_percent_as_number": 100 }
查看pending tasks：
GET /_cat/pending_tasks
其中 priority 字段则表示该 task 的优先级
查看分片未分配原因
GET _cluster/allocation/explain
其中reason 字段表示哪种原因导致的分片未分配，detail 表示详细未分配的原因
查看所有未分配的索引和主分片：
GET /_cat/indices?v&health=red
查看哪些分片出现异常
curl -s http://ip:port/_cat/shards | grep UNASSIGNED
重新分配一个主分片：
POST _cluster/reroute?pretty" -d '{ "commands" : [ { "allocate_stale_primary" : { "index" : "xxx", "shard" : 1, "node" : "12345...", "accept_data_loss": true } } ] }
其中node为es集群节点的id，可以通过curl ‘ip:port/_node/process?pretty’ 进行查询
降低索引的副本的数量
PUT /szv_ingress_*/settings { "index": { "number_of_replicas": 1 } }