4 自动发现SNMP OIDs

概述

在本章节中，我们将在交换机上执行一个SNMP 发现 。

从Zabbix server/proxy 6.4开始，支持 SNMP OIDs 这种发现方法。

监控项，触发器，图形原型

必须使用自动发现规则中的宏来创建 相关项目 类型监控项原型。

相关项目监控项从 walk[] 主监控项中获取值。因此，不需要让每个自动发现监控项单独再去轮询 SNMP 设备。

触发器和图形原型也可以通过使用自动发现规则中的宏来创建。.

Dynamic indexes with walk[]

Dynamic indexes (for example, interface indexes) can shift when hardware is reconfigured. To accommodate this behavior, a master SNMP walk discovery rule is created with a key such as:

walk[1.3.6.1.2.1.2.2.1.10]

After SNMP walk to JSON preprocessing, the result might resemble:

[
           {
               "{#SNMPINDEX}": "2",
               "{#VALUE}": "123456"
           },
           {
               "{#SNMPINDEX}": "3",
               "{#VALUE}": "654321"
           }
       ]

A dependent item prototype uses the {#SNMPINDEX} macro to construct the key:

net.if.in[{#SNMPINDEX}]

Preprocessing for this prototype includes "SNMP walk value" name with "1.3.6.1.2.1.2.2.1.10.{#SNMPINDEX}" OID in the Parameter field. Format: "Unchanged".

At runtime, actual items such as net.if.in[2] and net.if.in[3] are created. If a given interface index changes (for example, if the index 2 is replaced by 5 in the SNMP table), then on the next run of the discovery rule:

• The old dependent item net.if.in[2] is marked as "lost" or removed, and no new data is gathered for that item.
• A new dependent item net.if.in[5] is created, starting with an empty history.
• Historical data from net.if.in[2] is not automatically moved to net.if.in[5].

Trigger prototype example:

{Template_Interface:net.if.in[{#SNMPINDEX}].last()} > 1000000000

Graph prototype example includes items:

net.if.in[{#SNMPINDEX}]
       net.if.out[{#SNMPINDEX}]

This configuration ensures reliable monitoring of tables with dynamic indexes while minimizing SNMP traffic—only a single SNMP walk per poll cycle is required, with dependent item prototypes extracting the necessary values.

被发现的实体

当服务器运行时，它将根据 SNMP 发现规则返回的值创建真实的依赖项、触发器和图表。