Zabbix-monitoring system

# 一, zabbix monitoring architecture

The monitoring architecture of zabbix In the actual monitoring architecture, zabbix is ​​based on the network environment, monitoring scale and other architectures: server-client, server-proxy-client.

• server-clientThe architecture is also the simplest architecture of zabbix. No agent is passed between the monitoring machine and the monitored machine, and data is directly exchanged between the zabbix server and zabbix agentd. It is suitable for the monitoring environment where the network is relatively simple and the equipment is relatively few.
• server-proxy-clientIn the architecture, the proxy is a bridge between the server and the client. The proxy itself has no front end, and it does not store data itself. It only temporarily stores the data sent by agentd and submits it to the server. Generally applicable to the monitoring of medium-sized network architectures across computer rooms and across networks
• master-node-client

General monitoring:

[External chain image transfer failed, the source site may have an anti-theft chain mechanism, it is recommended to save the image and upload it directly (img-Zk1uviu3-1585731167452) (https://upload-images.jianshu.io/upload_images/21294643-1ce75294e8074125.png ? imageMogr2 / auto-orient / strip% 7CimageView2 / 2 / w / 1240)]

Distributed monitoring:

[External chain image transfer failed, the source site may have an anti-theft chain mechanism, it is recommended to save the image and upload it directly (img-ltem6Coi-1585731167453) (https://upload-images.jianshu.io/upload_images/21294643-2928810b2a20ba5e.png ? imageMogr2 / auto-orient / strip% 7CimageView2 / 2 / w / 1240)]

# 2, Zabbix advantages

• Open source, no software cost investment
• Server has low requirements for device performance
• Supports multiple devices and comes with multiple monitoring templates
• Support distributed centralized management, with automatic discovery function, can realize automatic monitoring
• When the monitored items are relatively large and the server queue is relatively large, the passive state can be adopted. The monitored client actively downloads the items to be monitored from the server and then uploads the data to the server. This method puts less load on the server.
• Api support, easy to integrate with other systems

# 3, Zabbix disadvantages

An agent needs to be installed on the monitored host. All data is stored in the database, and the generated data is very large. The bottleneck is mainly in the database.

# 四, Zabbix monitoring system monitoring objects

[External chain image transfer failed, the source site may have an anti-theft chain mechanism, it is recommended to save the image and upload it directly (img-I9EY7cFe-1585731167453) (https://upload-images.jianshu.io/upload_images/21294643-7bebcadc49ee9aa9.png ? imageMogr2 / auto-orient / strip% 7CimageView2 / 2 / w / 1240)]

• IPMI: Intelligent Platform Management Interface (Intelligent Platform Management Interface) IPMI can across different operating systems, firmware and hardware platforms, intelligent monitoring, control and automatic return of a large number of operational status of the server, the server system to reduce costs.

• SNMP: Network Management Protocol ( SNMP ) is a standard protocol specifically designed to manage network nodes ( servers , workstations , routers , switches, etc.) in an IP network . It is an application layer protocol.

# 五 、 Monitoring difference

• Nagios graphics are not particularly good, you can also install graphics plug-ins, but not very good-looking
• In general, nagios can be used with cacti if you need graphics
• Cacti's monitoring is polling monitoring, the efficiency is low, and the graphics are more beautiful than nagios
• Because zabbix and nagios are concurrent monitoring, they have higher requirements on cpu
• Zabbix is ​​much more powerful in performance and functionality
• Zabbix's graphics are pretty
• Support multiple monitoring methods zabbix-agent snmp, etc.
• Support distributed monitoring, there are many agents that can monitor
• Zabbix has a graphical web configuration interface, simple configuration
• zabbix supports auto discovery

# 六 、 Zabbix monitoring mode
Passive mode
Active mode
### 1. Server
Zabbix server is the core component of agent program to report system availability, system integrity and statistical data, and is the core storage of all configuration information, statistical information and operation data.

### 2. Database storage
All configuration information and data collected by Zabbix are stored in the database.

### 3. Web interface
In order to easily access Zabbix from any place and any platform, we provide a web-based Zabbix interface. This interface is part of Zabbix Server and usually runs on the same physical machine as Zabbix Server.

### 4. Proxy
Zabbix proxy can collect performance and availability data for Zabbix Server.

• Zabbix Proxy is one that can collect monitoring data from one or more monitored devices and send the information to Zabbix sever, which basically works for the server. All collected data is cached locally and then transmitted to the Zabbix server to which the proxy belongs.
• Deploying Proxy is optional, but may be very beneficial to spread the load of a single Zabbix sever. If only the proxy collects data, the process on the server will reduce CPU consumption and disk I / O load.
• Zabbix proxy is an ideal solution for centralized monitoring of remote areas, branches, and networks without local administrators.
• Zabbix proxy requires a separate database.

### 5. Agent monitoring agent
Zabbix agents are deployed on the monitoring target and can actively monitor local resources and applications, and report the collected data to Zabbix Server.

### 6. Data flow

In terms of monitoring, in order to create an item for collecting data, you must first create a host. In terms of alarms, create triggers in the monitoring items, and trigger trigger actions through triggers. Therefore, if you want to receive an alarm that Server XCPU is overloaded, you must:

1. Create a host for the Server and associate a monitoring item (Item) for monitoring the CPU.
2. Create a trigger Trigger, set to trigger when the CPU load is too high.
3. Trigger is triggered and an alarm email is sent.

# 七 、 The meaning of Zabbix common terms
### 1. Host

• A network device you want to monitor, expressed by IP or domain name

### 2. Host group

• A logical group of hosts; it contains hosts and templates. There is no direct relationship between hosts and templates in a host group. Host groups are usually used when assigning permissions to hosts in different user groups.

### 3. Item (item)

• The specific data of the host you want to receive, a metric data.

### 4. Trigger

• A logical expression used to define the problem threshold and "evaluate" the data received by the monitoring item
• When the received data is higher than the threshold, the trigger changes from "OK" to "Problem" state. When the received data is below the threshold, the trigger retains / returns an "OK" status.

### 5. Action

• A predefined action that reacts to the event.
• An action consists of an operation (e.g. issuing a notification) and a condition (the operation is occurring at the time)

### 6. Media

• Means of sending warning notices; ways of warning notices

### 7. Remote command

• A predefined command that can be automatically executed on the monitored host when certain conditions are met

### 8. Template

• A set of entities (monitoring items, triggers, graphics, applications, web scenes, etc.) that can be applied to one or more hosts
• The task of the template is to speed up the implementation of host monitoring tasks; it can also make batch modification of monitoring tasks easier. Templates are directly linked to each individual host.

### 9. Web scenario

• Use one or more HTTP requests to check website availability

### 10. Frontend

• Web interface provided by Zabbix

###11. Zabbix server

• Zabbix software realizes the core program of monitoring, the main function is to interact with Zabbix server and Agents, trigger calculation, send alarm notification; and save the data centrally, etc.

###12. Zabbix agent

• Zabbix agent is deployed on the monitoring target and actively monitors local resources and applications (hardware, memory, processor statistics, etc.).
• Zabbix agent collects local operation information and reports the data to Zabbix server for further processing. Once an exception occurs (such as full hard disk space or a crashed service process), Zabbix server will actively warn the administrator of the exception on the specified machine.