ES监控告警通过钉钉接口（python）

es自适应需求监控

#_*_coding:utf-8 _*_

from elasticsearch import Elasticsearch
import re
import time
from dingding import dd_info
import json
class Es_monitor():
    def __init__(self):
        ###es连接
        self.connect = Elasticsearch("http://xxx:9200",http_auth=('f', 'o'),port=9200)
    def es_status(self):
        mes = self.connect.cat.health()
        if re.search("red",mes):
            data = '{"content":{"es_cluster_status":"es集群状态为red"}}'
        elif re.search("yellow", mes):
            data = '{"content":{"es_cluster_status":"es集群状态为yellow"}}'
        dd_info(data)

    def pool(self):
        mes = self.connect.cat.nodes(h="n")
        nodesname = mes[0:-1].strip().split('\n')
        try:
            for node in nodesname:
                jsonData = self.connect.nodes.stats(node_id=node)
                nodeID = jsonData['nodes'].keys()
                search = jsonData['nodes'][list(nodeID)[0]]['thread_pool']['search']['threads']
                load = jsonData['nodes'][list(nodeID)[0]]['os']['cpu']['load_average']
                cpu = jsonData['nodes'][list(nodeID)[0]]['process']['cpu']['percent']
                #write = jsonData['nodes'][list(nodeID)[0]]['thread_pool']['write']['threads']
                if search > 100:
                    data = '{"content":{"es_search_pool":"%s节点es查询线程池大于100"}}'%(node)
                    dd_info(data)
                #elif load['1m'] > 20:
                #    data = '{"content":{"es_load":"%s节点负载[load(1m):%s,load(5m):%s]"}}'%(node,load['1m'],load['5m'])
                #    dd_info(data)
                elif cpu > 95:
                    data = '{"content":{"es_cpu":"%s节点cpu使用率为%s"}}'%(node,cpu)
                    dd_info(data)
        except:
            pass
    def es_search_rate(self):
        mes =self.connect.indices.stats()
        time.sleep(1)
        mes_2 = self.connect.indices.stats()
        search_result = ( mes_2['_all']['primaries']['search']['query_total']  -  mes['_all']['primaries']['search']['query_total'])
        #write_result = (mes_2['_all']['total']['indexing']['index_total'] - mes['_all']['total']['indexing']['index_total']) / 8
        order_search_rate = mes_2['indices']['-96-2020-2']['primaries']['search']['query_total'] - mes['indices']['-96-2020-2']['primaries']['search']['query_total']
        customer_search_rate = mes_2['indices']['tms-pro-customer-95']['primaries']['search']['query_total'] - mes['indices']['tms-pro-customer-95']['primaries']['search']['query_total']
        try:
            if search_result > 8000:
                data = '{"content":{"es_search_rate":"es查询qps大于8000,当前值:%s/s"}}'%(search_result)
                dd_info(data)
            elif order_search_rate > 1300:
                data = '{"content":{"es_search_rate":"Order索引查询qps大于1300,当前值:%s/s"}}'%(order_search_rate)
                dd_info(data)
            elif customer_search_rate> 1300:
                data = '{"content":{"es_search_rate":"Customer索引查询qps大于800,当前值:%s/s"}}'%(customer_search_rate)
                dd_info(data)
        except:
            pass
    def main(self):
        #self.es_status()
        self.pool()
        self.es_search_rate()

if __name__ == '__main__':
    ES = Es_monitor()
    ##es集群状态
    ES.main()

钉钉告警

#!/usr/bin/env python
#-*- coding: utf-8 -*-
import os
import sys
import json
import datetime
import requests

def dd_info(data):
    url = 'https://oapi.dingtalk.com/robot/'
    #url = 'https://open.feishu.cn/open-apis/bot/'
    headers = {
        'Content-Type': 'application/json;charset=utf-8'
    }
    formdata = {
        "msgtype": "text",
        "text": {"content": data
        }
    }
    mes = requests.post(url=url, data=json.dumps(formdata), headers=headers)
    mes.text
if __name__ == '__main__':
   dd_info()

查询索引监控

import os
import time
import json

def mes(index):
    message = os.popen('curl -s -u xxx:xxx -XGET "http://xxxxx:9200/%s/_stats"'%(index))
    m = message.read()
    message.close()
    messages = json.loads(m)
    mes_qps = messages['_all']['primaries']['search']['query_total']
    mes_millis = messages['_all']['primaries']['search']['query_time_in_millis']
    return mes_qps,mes_millis

def read():
    while True:
        ###查询值存入字典，做对比。
        values = dict()
        values['qps'] = mes()[0]
        values['millis'] = mes()[1]
        time.sleep(10)
        ###单次索引查询，9为分片数
        if int(mes()[0]) - 9 == int(values['qps']):
            value = (int(mes()[1]) - int(values['millis'])) / 9
            print(value)
        ###查询无延迟
        elif int(values['qps']) == int(mes()[0]):
            value = int(mes()[1]) - int(values['millis'])
            print(value)
        ###均次查询
        else:
            value = (int(mes()[1]) - int(values['millis'])) / (int(mes()[0]) - int(values['qps']))
        print(value)
        #else:
        #    num = (int(mes()[0]) - int(values['qps']))
        #    values = int(mes()[1]) - int(values['millis'])
        #    value= int(values) / int(num)

if __name__ == '__main__':
    read(‘xxxxx’)

写入数据监控

import os
import time
import json

def mes():
    message = os.popen('curl -s -u xxx:xxxx -XGET "http://xxxxx:9200/xxxxxxxx/_stats/indexing"')
    m = message.read()
    message.close()
    messages = json.loads(m)
    mes_write = messages['_all']['primaries']['indexing']['index_total']
    return mes_write

def write():
    while True:
        ###查询值存入字典，做对比。
        values = dict()
        values['write'] = mes()
        time.sleep(10)
        ###索引写入计算
        value = (int(mes())-int(values['write']))/9
        print(value)
if __name__ == '__main__':
    write()