阿里云starrocks监控告发至钉钉群

背景:新入职一家公司,现场没有对sr的进行监控,根据开发的需求编写了一个python脚本。

脚本逻辑:抓取sr的be/fe/routine load状态信息,判读是否触发告警,若满足告警条件,则发送告警信息到钉钉群,并艾特对应的责任人。

python 复制代码
# -*- coding: utf-8 -*-
""""
author: zjh
date: 2023-09-28
description:StarRocks cluster monitoring, include routine load,frontend,backend status.
             And then send warn message to dingtalk.
"""
import base64
import urllib
import requests
import json
import time
import hmac
import hashlib
import pymysql


def msg(warntype, content, token):
    # 根据实际修改1:钉钉机器人地址
    dingding_url = 'https://oapi.dingtalk.com/robot/send?access_token=' + token
    # 根据实际修改2:钉钉的加签
    secret = 'SECd0c0f14733789a******************645a4d1ae76ea3481b1384f5ef'

    # 加签算法,钉钉文档里的
    timestamp = str(round(time.time() * 1000))
    secret_enc = secret.encode('utf-8')
    string_to_sign = '{}\n{}'.format(timestamp, secret)
    string_to_sign_enc = string_to_sign.encode('utf-8')
    hmac_code = hmac.new(secret_enc, string_to_sign_enc, digestmod=hashlib.sha256).digest()
    sign = urllib.parse.quote_plus(base64.b64encode(hmac_code))

    headers = {'Content-Type': 'application/json;charset=utf-8'}
    api_url = dingding_url + "&timestamp={}&sign={}".format(timestamp, sign) ##加签算法使用

    if warntype == 'routineloadwarn':
        owner=["dfdfs134"]

    if warntype == 'fewarn' or warntype == 'bewarn':
        owner=["faasdfaAQE"]

    json_text = {
        "at": {
            "atMobiles": [
                "180xxxxxx"
            ],
            "atDingtalkIds":owner,
            "isAtAll": "False"
        },
        "msgtype": "text",
        "text": {
            "content": content
        }

    }

    requests.post(api_url, json.dumps(json_text), headers=headers)

def srmsg(ip,port,username,passwd,db,exe_sql):
    conn = pymysql.connect(host=ip,port=port,user=username,password=passwd,database=db)
    cursor = conn.cursor()
    sql = exe_sql
    cursor.execute(sql)
    result=cursor.fetchall()
    cursor.close()
    conn.close()
    return  result

def fewarn(clustername,base_tuple_fe,token):
    warntype='fewarn'
     #2.4.1版本SR
    if clustername == 'SR':
        for feinfo in base_tuple_fe:
            # print(feinfo)
            if feinfo[10] == "false":
                fewanrmsg = "集群: " + clustername + "\n" + \
                            "异常FE: " + feinfo[1] + "\n" + \
                            "状态: " + feinfo[10] + "\n" + \
                            "上次启动时间: " + feinfo[15]
                msg(warntype, fewanrmsg, token)
    else:
     #2.4.1版本SR
        for feinfo in base_tuple_fe:
            # print(feinfo)
            if feinfo[9] == "false":
                fewanrmsg = "集群: " + clustername + "\n" + \
                            "异常FE: " + feinfo[1] + "\n" + \
                            "状态: " + feinfo[9] + "\n" + \
                            "上次启动时间: " + feinfo[14]
                msg(warntype, fewanrmsg, token)

def bewarn(clustername,base_tuple_be,token):
    warntype='bewarn'
    #2.4.1版本SR
    if clustername == 'SatrRocks':
        for beinfo in base_tuple_be:
            # print(beinfo)
            if beinfo[9] == "false":
                bewanrmsg = "集群: " + clustername + "\n" + \
                            "异常BE: " + beinfo[2] + "\n" + \
                            "状态: " + beinfo[9] + "\n" + \
                            "上次启动时间: " + beinfo[7]
                msg(warntype, bewanrmsg, token)
    else:
    #2.3.1版本sr
        for beinfo in base_tuple_be:
            # print(beinfo)
            if beinfo[8] == "false":
                bewanrmsg = "集群: " + clustername + "\n" + \
                            "异常BE: " + beinfo[1] + "\n" + \
                            "状态: " + beinfo[8] + "\n" + \
                            "上次启动时间: " + beinfo[6]
                msg(warntype, bewanrmsg, token)

def routineloadwarn(clustername,load_tuple,token):
    warntype='routineloadwarn'
    for info in load_tuple:
        #title = dict_srprod["envname"]
        input_message= "集群: "+ clustername + "\n" +\
                       "任务id: "+ info[0] + "\n" +\
                       "任务名: "+ info[1] + "\n" +\
                       "库名: " + info[5]+ "\n" +\
                       "表名: " + info[6]+ "\n" +\
                       "状态: " + info[7]+ "\n" +\
                       "挂起时间: "+ info[2] + "\n" +\
                       "日志链接:" + info[16] + "\n" +\
                       "其他错误:" + info[17]
        msg(warntype, input_message, token)



if __name__ == '__main__':
    dict_srprod={"envname":"生产SatrRocks","ip":"192.168.10.10","port":29030,"username":"root","passwd":"********","dbname":["test1","test2"],
                 "check_fe_status":"show frontends;","check_be_status":"show backends;","check_rtload":"show routine load where State='PAUSED' and TableName in ('t1','t2');"
                 }
    dict_qwprod={"envname":"生产(1)SatrRocks","ip":"192.168.10.10","port":39030,"username":"root","passwd":"********","dbname":["test0","test"],
                 "check_fe_status":"show frontends;","check_be_status":"show backends;","check_rtload":"show routine load where State='PAUSED';"
                 }
    dict_srsit={"envname":"测试SatrRocks","ip":"192.168.10.11","port":19030,"username":"root","passwd":"********","dbname":["test1","test2","test3"],
                 "check_fe_status":"show frontends;","check_be_status":"show backends;","check_rtload":"show routine load where State='PAUSED' and TableName='t0';"
                 }

    token = "xxxxx"

    """
   环境0
    """
    #生产 routine load monitor
    base_tuple=srmsg(dict_srprod["ip"],dict_srprod["port"],
          dict_srprod["username"],dict_srprod["passwd"],
          dict_srprod["dbname"][0],dict_srprod["check_rtload"])

    routineloadwarn(dict_srprod["envname"],base_tuple,token)

    # 生产be/fe状态监控
    base_tuple_fe = srmsg(dict_srprod["ip"], dict_srprod["port"],
                          dict_srprod["username"], dict_srprod["passwd"],
                          dict_srprod["dbname"][0], dict_srprod["check_fe_status"])
    fewarn(dict_srprod["envname"],base_tuple_fe,token)

    base_tuple_be = srmsg(dict_srprod["ip"], dict_srprod["port"],
                          dict_srprod["username"], dict_srprod["passwd"],
                          dict_srprod["dbname"][0], dict_srprod["check_be_status"])
    bewarn(dict_srprod["envname"],base_tuple_be,token)

    """
    环境1
    """
    #### routine load monitor
    base_tuple=srmsg(dict_qwprod["ip"],dict_qwprod["port"],
          dict_qwprod["username"],dict_qwprod["passwd"],
          dict_qwprod["dbname"][0],dict_qwprod["check_rtload"])

    routineloadwarn(dict_qwprod["envname"],base_tuple,token)

    # be/fe状态监控
    base_tuple_fe = srmsg(dict_qwprod["ip"], dict_qwprod["port"],
                          dict_qwprod["username"], dict_qwprod["passwd"],
                          dict_qwprod["dbname"][0], dict_qwprod["check_fe_status"])
    fewarn(dict_qwprod["envname"],base_tuple_fe,token)

    base_tuple_be = srmsg(dict_qwprod["ip"], dict_qwprod["port"],
                          dict_qwprod["username"], dict_qwprod["passwd"],
                          dict_qwprod["dbname"][0], dict_qwprod["check_be_status"])
    bewarn(dict_qwprod["envname"],base_tuple_be,token)

    """
    环境2
    """
    ####测试routine load monitor
    base_tuple=srmsg(dict_srsit["ip"],dict_srsit["port"],
          dict_srsit["username"],dict_srsit["passwd"],
          dict_srsit["dbname"][0],dict_srsit["check_rtload"])

    routineloadwarn(dict_srsit["envname"],base_tuple,token)

    # 测试be/fe状态监控
    base_tuple_fe = srmsg(dict_srsit["ip"], dict_srsit["port"],
                          dict_srsit["username"], dict_srsit["passwd"],
                          dict_srsit["dbname"][0], dict_srsit["check_fe_status"])
    fewarn(dict_srsit["envname"],base_tuple_fe,token)

    base_tuple_be = srmsg(dict_srsit["ip"], dict_srsit["port"],
                          dict_srsit["username"], dict_srsit["passwd"],
                          dict_srsit["dbname"][0], dict_srsit["check_be_status"])
    bewarn(dict_srsit["envname"],base_tuple_be,token)

另外,钉钉的一些开发说明,请参阅钉钉官网

注册机器人链接:https://open.dingtalk.com/document/connector/alarm-subscription

相关推荐
阿里云云原生16 分钟前
Android App 崩溃排查实战:如何利用 RUM 完整数据与符号化技术定位问题?
android·阿里云·云原生·rum
有想法的py工程师17 分钟前
PostgreSQL + Debezium CDC 踩坑总结
数据库·postgresql
Nandeska31 分钟前
2、数据库的索引与底层数据结构
数据结构·数据库
小卒过河01041 小时前
使用apache nifi 从数据库文件表路径拉取远程文件至远程服务器目的地址
运维·服务器·数据库
过期动态1 小时前
JDBC高级篇:优化、封装与事务全流程指南
android·java·开发语言·数据库·python·mysql
Mr.朱鹏1 小时前
SQL深度分页问题案例实战
java·数据库·spring boot·sql·spring·spring cloud·kafka
一位代码1 小时前
mysql | 常见日期函数使用及格式转换方法
数据库·mysql
SelectDB1 小时前
Apache Doris 4.0.2 版本正式发布
数据库·人工智能
杰克尼2 小时前
mysql_day01
数据库·mysql
ccino .2 小时前
sql注入中过滤分隔符的测试方法
数据库·sql