linux监控elasticsearch,zabbix4.2 监控Elasticsearch集群指标

监控Elasticsearch

Elasticsearch监控原理

Elasticsearch是一个基于Lucene的搜索服务器，它提供了一个具有分布式多用户能力的全文搜索引擎，基于RESTful Web接口。Elasticsearch是使用Java语言开发的，对于监控数据，其本身提供了API接口供外部获取数据。使用如下命令获取监控指标数据：

curl “http://localhost:9200/_cluster/health”) 集群的健康情况

http://192.168.6.21:9200/_cluster/state?pretty 查看集群的状态信息

curl “http://localhost:9200/_nodes/_local/stats?all=true”

获取数据之后，我们编写一些代码将数据输出为zabbix需要的格式。

关于Elasticsearch的监控指标，官方文档地址：

响应的内容解释

内容

解释

"cluster_name" : "my-elk-cluster",

集群名

"status" : "green",

集群健康状态，正常的话是“green”，缺少副本分片为“yellow”，缺少主分片为“red”

"timed_out" : false,

"number_of_nodes" : 1,

集群节点数

"number_of_data_nodes" : 1,

数据节点数

"active_primary_shards" : 0,

主分片数

"active_shards" : 0,

可用的分片数

"relocating_shards" : 0,

正在迁移的分片数

"initializing_shards" : 0,

正在初始化的分片数

"unassigned_shards" : 0,

未分配的分片，但在集群中存在

"delayed_unassigned_shards" : 0,

延时待分配到具体节点上的分片数

"number_of_pending_tasks" : 0,

待处理的任务数，指主节点创建索引并分配shards等任务

"number_of_in_flight_fetch" : 0,

"task_max_waiting_in_queue_millis" : 0,

"active_shards_percent_as_number" : 100.0

可用分片数占总分片的比例

Elasticsearch监控脚本

放在/etc/zabbix/scripts/或/usr/local/zabbix/scripts

key_elasticsearch.sh

#!/bin/bash

case $1 in

cluster_name)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F" '/cluster_name/ {print $4}' ;;

status)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F" 'NR==3 {print $4}' ;;

timed_out)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==4 {print $1}' |awk -F: '{print $2}' ;;

number_nodes)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==5 {print $1}' |awk -F: '{print $2}' ;;

data_nodes)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==6 {print $1}' |awk -F: '{print $2}' ;;

active_primary_shards)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==7 {print $1}' |awk -F: '{print $2}' ;;

active_shards)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==8 {print $1}' |awk -F: '{print $2}' ;;

relocating_shards)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==9 {print $1}' |awk -F: '{print $2}' ;;

initializing_shards)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==10 {print $1}' |awk -F: '{print $2}' ;;

unassigned_shards)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==11 {print $1}' |awk -F: '{print $2}' ;;

delayed_unassigned_shards)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==12 {print $1}' |awk -F: '{print $2}' ;;

number_of_pending_tasks)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==13 {print $1}' |awk -F: '{print $2}' ;;

active_shards_percent_as_number)

curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==16 {print $1}' |awk -F: '{print $2}' ;;

*)

echo "Usage: $0 { cluster_name | status | timed_out | number_nodes | data_nodes | active_primary_shards | active_shards | relocating_shards | initializing_shards | unassigned_shards|delayed_unassigned_shards|number_of_pending_tasks|active_shards_percent_as_number}" ;;

esac

在shell脚本里，“curl -s -XGET 'http://192.168.6.21:9200/_cluster/health?pretty' |awk -F, 'NR==16 {print $1}' |awk -F: '{print $2}''”这样的命令，“NR==16”是指在浏览器访问http://192.168.6.21:9200/_cluster/health?pretty，获取页面的第16行(从第1行的“{”开始计数)。

脚本授予执行权限

chmod +x /etc/zabbix/scripts/*.sh

属主、属组可能也需要授权

chown zabbix:zabbix /etc/zabbix/scripts/*.sh

配置文件

目录/etc/zabbix/zabbix_agentd.d/userparameter_rabbitmq.conf或/usr/local/zabbix/etc/zabbix_agentd.conf.d/userparameter_rabbitmq.conf

UserParameter=elasticsearch.cluster_health[*],/etc/zabbix/scripts/key_elasticsearch.sh $1 $2

重启zabbix_agent

systemctl restart zabbix_agentd

yum安装的zabbix-agent

systemctl restart zabbix-agent.service

测试获取RadditMQ数据

zabbix_get -s 192.168.6.21 -k elasticsearch.cluster_health.active_primary_shards

Elasticsearch监控指标

应用集：Elasticsearch

名称

键值

Elasticsearch cluster active primary shards

elasticsearch.cluster_health[active_primary_shards]

Elasticsearch cluster active shards

elasticsearch.cluster_health[active_shards]

Elasticsearch cluster active shards percent

elasticsearch.cluster_health[active_shards_percent]

Elasticsearch cluster delayed unassigned shards

elasticsearch.cluster_health[delayed_unassigned_shards]

Elasticsearch cluster health

elasticsearch.cluster_health[status]

Elasticsearch cluster health int

elasticsearch.cluster_health[status_int]

Elasticsearch cluster initializing shards

elasticsearch.cluster_health[initializing_shards]

Elasticsearch cluster name

elasticsearch.cluster_health[cluster_name]

Elasticsearch cluster number of data nodes

elasticsearch.cluster_health[number_of_data_nodes]

Elasticsearch cluster number of in flight fetch

elasticsearch.cluster_health[number_of_in_flight_fetch]

Elasticsearch cluster number of nodes

elasticsearch.cluster_health[number_of_nodes]

Elasticsearch cluster number of pending tasks

elasticsearch.cluster_health[number_of_pending_tasks]

Elasticsearch cluster relocating shards

elasticsearch.cluster_health[relocating_shards]

Elasticsearch cluster task max waiting in queue

elasticsearch.cluster_health[task_max_waiting_in_queue_millis]

Elasticsearch cluster timeout

elasticsearch.cluster_health[timed_out]

Elasticsearch cluster unassigned shards

elasticsearch.cluster_health[unassigned_shards]

Elasticsearch get stats

elasticsearch.stats[{$ZABBIX_SERVER_IP},{$ES_ADDRESS},{$ES_ZBX_PREFIX},{$ES_USER},{$ES_PASSWORD},{$CA_PATH},{$ZABBIX_HOSTNAME}]

Elasticsearch port listen

net.tcp.service[tcp,,{$ES_PORT}]

Elasticsearch process num

proc.num[,,,bootstrap.Elasticsearch]

Elasticsearch监控触发器