zabbix报警逻辑初探

时间 2019-11-10

原文原文链接

zabbix报警逻辑初探

首先贴出一张网上找的一张关于zabbix报警相关表结构及表关联逻辑图：
前端

actions表

actions表对应前端配置是动做(actions)
数据库

action由condition(条件)和operations(操做)组成。当知足指定的条件，而后执行操做。发送内容在动做里配置(好比默认状况下没有加上报警产生时间，能够人为加上去)。网络

MariaDB [rtm]> desc actions;
+---------------+---------------------+------+-----+---------+-------+
| Field         | Type                | Null | Key | Default | Extra |
+---------------+---------------------+------+-----+---------+-------+
| actionid      | bigint(20) unsigned | NO   | PRI | NULL    |       |
| name          | varchar(255)        | NO   | UNI |         |       |
| eventsource   | int(11)             | NO   | MUL | 0       |       |
| evaltype      | int(11)             | NO   |     | 0       |       |
| status        | int(11)             | NO   |     | 0       |       |
| esc_period    | int(11)             | NO   |     | 0       |       |
| def_shortdata | varchar(255)        | NO   |     |         |       |
| def_longdata  | text                | NO   |     | NULL    |       |
| recovery_msg  | int(11)             | NO   |     | 0       |       |
| r_shortdata   | varchar(255)        | NO   |     |         |       |
| r_longdata    | text                | NO   |     | NULL    |       |
| formula       | varchar(255)        | NO   |     |         |       |
+---------------+---------------------+------+-----+---------+-------+

actionid:   action的id
name：   action的名称
eventsource：    对应events表的source
evaltype:   conditions里计算方式的id(0:and/or, 1:and, 2:or, 3:自定义表达式)
status:     启用状态(0启用， 1禁用)
esc_period:     步骤操做持续时间
def_shortdata:      subject
def_longdata:       message内容
recovery_msg:       恢复消息启用状态，1-启用
r_shortdata：    恢复subject
r_longdata      恢复message
fromula:        conditions里自定义的表达式

zabbix事件

zabbix事件一共有三种，分别为：触发器事件、发现事件、内部事件、自动注册事件。3d

zabbix内部事件日志
- 监控项item状态从normal变为unsupported，或者从unsupported变为normal
- low-level发现规则状态从normal变为unsupported，或者从unsupported变为normal
- 触发器状态从normal变为unknown，或者从unknown变为normal
zabbix发现事件code
- 配置网络发现规则以后，zabbix会按期按照这个规则去扫描IP段，一旦发现主机和服务，便生成一个事件
- zabbix自动发现
zabbix触发事件orm

触发器状态发生变化生成一个包含详细状态信息的触发器事件server
zabbix自动注册事件blog

active agent主动与server通讯，zabbix server使用agent通讯的ip地址与端口来添加主机，并生成一个自动注册事件队列

events表

MariaDB [rtm]> select * from events where source=0;
+---------+--------+--------+----------+------------+-------+--------------+-----------+--------------------------------------------+
| eventid | source | object | objectid | clock      | value | acknowledged | ns        | description                                |
+---------+--------+--------+----------+------------+-------+--------------+-----------+--------------------------------------------+
|     317 |      0 |      0 |    13075 | 1548827260 |     0 |            0 | 399008160 | 99.9512<5                                  |
|     318 |      0 |      0 |    13467 | 1548827312 |     0 |            0 | 696464358 | (0=0 and 0.1854>75) or (0=1 and 0.1854>65) |
|     308 |      0 |      0 |    13468 | 1548827253 |     0 |            1 | 367035016 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     309 |      0 |      0 |    13469 | 1548827254 |     0 |            0 | 352296205 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     310 |      0 |      0 |    13470 | 1548827255 |     0 |            0 | 363172506 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     311 |      0 |      0 |    13471 | 1548827256 |     0 |            0 | 375124809 | (0=0 and 0.0169>75) or (0=1 and 0.0169>65) |
|     319 |      0 |      0 |    13472 | 1548827257 |     0 |            0 | 373863748 | (0=0 and 2.5554>75) or (0=1 and 2.5554>65) |
|     320 |      0 |      0 |    13473 | 1548827258 |     0 |            0 | 381757318 | (0=0 and 0.0846>75) or (0=1 and 0.0846>65) |
|     321 |      0 |      0 |    13474 | 1548827259 |     0 |            0 | 388674314 | (0=0 and 0.2199>75) or (0=1 and 0.2199>65) |
|     322 |      0 |      0 |    13475 | 1548827260 |     0 |            0 | 398635590 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     323 |      0 |      0 |    13479 | 1548827264 |     0 |            0 | 425321837 | (0=0 and 3.1495>75) or (0=1 and 3.1495>65) |
|     324 |      0 |      0 |    13480 | 1548827265 |     0 |            0 | 429536321 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     325 |      0 |      0 |    13481 | 1548827266 |     0 |            0 | 439574519 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     326 |      0 |      0 |    13482 | 1548827267 |     0 |            0 | 441541684 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     327 |      0 |      0 |    13483 | 1548827268 |     0 |            0 | 448121449 | (0=0 and 0>75) or (0=1 and 0>65)           |
|     328 |      0 |      0 |    13484 | 1548827269 |     0 |            1 | 460702185 | (0=0 and 0.0406>75) or (0=1 and 0.0406>65) |


action里可根据事件源(四个事件类型)建立不一样的动做。和这里的source是对应起来的。
objectid对应的是triggers表里的triggerid
value等于0就是OK，等于1就是PROBLEM
acknowledged=0就是未确认，等于1就是已确认

source=0的就是触发器事件
source=1的就是自动发现事件
source=2的就是自动注册事件
source=3的就是内部事件

zabbix报警媒介自定义

zabbix媒介类型包括mail、sms、自定义脚本。

media_type表

MariaDB [rtm]> select * from media_type\G;
*************************** 1. row ***************************
        mediatypeid: 1
               type: 0
        description: Email
        smtp_server: mail.company.com
          smtp_helo: company.com
         smtp_email: rtm@company.com
          exec_path: 
          gsm_modem: 
           username: 
             passwd: 
             status: 0
          smtp_port: 25
      smtp_security: 0
   smtp_verify_peer: 0
   smtp_verify_host: 0
smtp_authentication: 0
        exec_params: 
*************************** 2. row ***************************
        mediatypeid: 2
               type: 3
        description: Jabber
        smtp_server: 
          smtp_helo: 
         smtp_email: 
          exec_path: 
          gsm_modem: 
           username: jabber@company.com
             passwd: rtm
             status: 0
          smtp_port: 25
      smtp_security: 0
   smtp_verify_peer: 0
   smtp_verify_host: 0
smtp_authentication: 0
        exec_params: 
*************************** 3. row ***************************
        mediatypeid: 3
               type: 2
        description: SMS
        smtp_server: 
          smtp_helo: 
         smtp_email: 
          exec_path: 
          gsm_modem: /dev/ttyS0
           username: 
             passwd: 
             status: 0
          smtp_port: 25
      smtp_security: 0
   smtp_verify_peer: 0
   smtp_verify_host: 0
smtp_authentication: 0
        exec_params: 
*************************** 4. row ***************************
        mediatypeid: 4
               type: 1
        description: 智能告警
        smtp_server: 
          smtp_helo: 
         smtp_email: 
          exec_path: sr_event/sr_event_client/sr_event_client.py
          gsm_modem: 
           username: 
             passwd: 
             status: 0
          smtp_port: 25
      smtp_security: 0
   smtp_verify_peer: 0
   smtp_verify_host: 0
smtp_authentication: 0
        exec_params: {ALERT.SUBJECT}\n
4 rows in set (0.00 sec)

media表

MariaDB [rtm]> desc media;
+-------------+---------------------+------+-----+-----------------+-------+
| Field       | Type                | Null | Key | Default         | Extra |
+-------------+---------------------+------+-----+-----------------+-------+
| mediaid     | bigint(20) unsigned | NO   | PRI | NULL            |       |
| userid      | bigint(20) unsigned | NO   | MUL | NULL            |       |
| mediatypeid | bigint(20) unsigned | NO   | MUL | NULL            |       |
| sendto      | varchar(100)        | NO   |     |                 |       |
| active      | int(11)             | NO   |     | 0               |       |
| severity    | int(11)             | NO   |     | 63              |       |
| period      | varchar(100)        | NO   |     | 1-7,00:00-24:00 |       |
+-------------+---------------------+------+-----+-----------------+-------+

media表数据来自用户配置的报警媒介。

alerts表

MariaDB [rtm]> show create table alerts\G;
*************************** 1. row ***************************
       Table: alerts
Create Table: CREATE TABLE `alerts` (
  `alertid` bigint(20) unsigned NOT NULL,
  `actionid` bigint(20) unsigned NOT NULL,
  `eventid` bigint(20) unsigned NOT NULL,
  `userid` bigint(20) unsigned DEFAULT NULL,
  `clock` int(11) NOT NULL DEFAULT '0',
  `mediatypeid` bigint(20) unsigned DEFAULT NULL,
  `sendto` varchar(100) COLLATE utf8_bin NOT NULL DEFAULT '',
  `subject` varchar(255) COLLATE utf8_bin NOT NULL DEFAULT '',
  `message` text COLLATE utf8_bin NOT NULL,
  `status` int(11) NOT NULL DEFAULT '0',
  `retries` int(11) NOT NULL DEFAULT '0',
  `error` varchar(128) COLLATE utf8_bin NOT NULL DEFAULT '',
  `esc_step` int(11) NOT NULL DEFAULT '0',
  `alerttype` int(11) NOT NULL DEFAULT '0',
  PRIMARY KEY (`alertid`),
  KEY `alerts_1` (`actionid`),
  KEY `alerts_2` (`clock`),
  KEY `alerts_3` (`eventid`),
  KEY `alerts_4` (`status`,`retries`),
  KEY `alerts_5` (`mediatypeid`),
  KEY `alerts_6` (`userid`),
  CONSTRAINT `c_alerts_1` FOREIGN KEY (`actionid`) REFERENCES `actions` (`actionid`) ON DELETE CASCADE,
  CONSTRAINT `c_alerts_2` FOREIGN KEY (`eventid`) REFERENCES `events` (`eventid`) ON DELETE CASCADE,
  CONSTRAINT `c_alerts_3` FOREIGN KEY (`userid`) REFERENCES `users` (`userid`) ON DELETE CASCADE,
  CONSTRAINT `c_alerts_4` FOREIGN KEY (`mediatypeid`) REFERENCES `media_type` (`mediatypeid`) ON DELETE CASCADE
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_bin
1 row in set (0.00 sec)

列出这个表结构是由于遇到过zabbix邮件队列阻塞，解决办法网上不多有资料，所以本身去后台查看了表信息。网上说有人经过删数据库数据解决。并且zabbix界面的动做日志数据都来自这张表。

删数据解决我没尝试过，我经过引流方式将原来阻塞的邮件方式转换为mediatypeid为自定义脚本操做将阻塞队列排空。

不过下来研究过表结构和事件逻辑，我以为能够这样尝试解决：

备份events和alerts表
查找出阻塞这段时间的events，根据这个eventsid去删除events相关记录和alerts相关记录。

还有一种方法就是在zabbix的general下的管家下面设置事件和报警相关数据存储时间为一天，这样其实也是经过删除数据库数据解决，不过是zabbix管家主动帮咱们作了。

时间仓促，后面再补充。。。。。。