Function introduction and common terms of ZABBIX

Zabbix Function introduction and common terms of

1. Zabbix Introduce

Zabbix It is an enterprise level distributed open source monitoring solution .

Zabbix It is a software that can monitor the health and integrity of many network parameters and servers .Zabbix Use flexible notification mechanisms , Allows users to configure mail based alerts for almost any event . This can quickly respond to server problems .Zabbix Excellent reporting and data visualization based on stored data . These functions make Zabbix Ideal for capacity planning .

Zabbix Support polling and passive capture . be-all Zabbix The report 、 Both statistics and configuration parameters can be obtained by Web Access the front page of . After proper configuration ,Zabbix You can monitor IT Infrastructure plays an important role .

2. Zabbix function

Data collection

• Availability and performance collection ;
• Support SNMP（ Including active polling and passive capture ）、IPMI、JMX、VMware monitor ;
• Custom check ;
• Collect the required data according to the user-defined time interval ;
• adopt Server/Proxy and Agents To perform data collection .

Flexible threshold definition

• Very flexible alarm thresholds can be defined , It's called a trigger , The trigger obtains the reference value from the back-end database .

Highly configurable alarms

• According to the incremental plan 、 The receiver 、 Send alarm notification by media type customization ;
• Using macro variables can make alarm notification more efficient and beneficial ;
• Automatic actions include remote commands .

Real time graphics

• Use the built-in graphics function to draw the monitoring items into graphics in real time .

Web Monitoring function

• Zabbix Can track the simulation mouse in Web Click operations on the website , To check Web The function and response time of the website .

Rich visualization options

• Be able to create custom graphics that can combine multiple monitoring items into a single view ;
• Network topology ;
• Display custom aggregate graphics and slide presentations in dashboard style ;
• report form ;
• High level of monitoring resources （ Business ） View .

Historical data storage

• Data stored in the database ;
• Configurable historical data ;
• Built in data management mechanism （housekeeping）.

Simple configuration

• Add the monitored device as the host ;
• Once the host is added to the database , Will collect host data for monitoring ;
• Use the template for monitoring equipment .

Apply template

• Group check in template ;
• Templates can be associated with other templates , Get inheritance .

Internet discovery

• Autodiscover network devices ;
• Zabbix Agent Automatically register after discovering the device ;
• Automatically discover file systems 、 Network interface and SNMP OIDs value .

shortcut Web Interface

• be based on PHP Of Web front end ;
• You can visit... From anywhere ;
• You can customize your operation ;
• The audit log .

Zabbix API

• Zabbix API by Zabbix Provide programmable interface , For batch operation 、 Third party software integration and other uses .

Authority management system

• Secure user authentication ;
• Restrict specific users to specific views .

Powerful and easy to expand Zabbix Agent

• Deployed on monitored objects ;
• Perfect support Linux and Windows ;

Binary daemons

• For better performance and less memory , use C Language writing ;
• Easy to transplant .

Adapt to more complex environments

• Use Zabbix Proxy agent , It can easily realize distributed remote monitoring .

3. zabbix Common monitoring parameters

Whether it's zabbix, Or other open source monitoring , At the end of the day, they're all doing five things ：

1. Data collection
2. Store the collected data
3. Analyze the stored data
4. Use icons to show the results of the analysis
5. Use various ways to alert the problem .

zabbix Common monitoring parameters

1. monitor web service

• web Whether the service is normal
• Business （ Whether the web page can be accessed 、 Whether the order can be completed 、 Registered users ）
• Service response time
• The concurrency of the service （ Active users 、 Inactive users ）

2. Monitoring the database

• Monitor disk usage
• Monitor memory usage
• View the number of concurrent connections
• Check the frequency of adding, deleting and modifying the database
• Check the master-slave status
• Check the database backup

3. Server monitoring

• disk
  • Usage rate
  • inode Count
  • block Count
  • Literacy rate
• CPU
  • monitor cpu load
  • Monitoring use cpu The process with the most resources
• Memory
  • Usage rate
  • buffer
  • Buffer zone
  • Swap partition size
• The Internet
  • Monitor the up look ahead rate of each network card
  • Monitor the processes that occupy more network bandwidth
  • Monitor packet loss
  • Monitor the blocking of network packets
• process
  • The total number of processes in the current system
  • Monitors the number of processes in a particular program

4. Classification of monitoring

1. Hardware monitoring —— Router 、 Switch 、 A firewall
2. System monitoring ——cpu、 Memory 、 disk 、 The Internet 、 process 、tcp
3. Service monitoring ——nginx、php、tomcat、redis、memcache、mysql
4.web monitor —— response time 、 Loading time 、 render time 、 Is the page 200
5. Log monitoring ——ELK、（ collect 、 Storage 、 analysis 、 Exhibition ） journal
6. Safety monitoring ——firewalld、WAF(nginx+lua)、 Safety treasure 、 Niudunyun 、 Safe dog

5. Zabbix Common terminology

• host （host）： Network devices to monitor , May by IP or DNS Name designation .
• Host group （host group）： The logical container of the host , Can contain hosts and templates , However, hosts and templates in the same group cannot be linked to each other , Host groups are usually used when assigning monitoring rights to users or user groups .
• Monitoring item （item）： Data related to a specific monitoring indicator , These data come from the monitored objects ,item yes zabbix The core of data collection , No, item, There's no data , Every item All by “key” Are identified .
• trigger （trigger）： An expression , Used to evaluate a specific... Of a monitored object item Whether the data received within the reasonable range , That's the threshold ; When the amount of data received is greater than the threshold , The trigger state will be from “OK” Turn into “Problem”, When the data volume returns to a reasonable range again , Its status will change from “Problem” Convert back to “OK”.
• event （event）： That is, an event worthy of attention , Such as the state transition of the trigger , new agent Or go back online agent Automatic registration, etc .
• action （action）： It refers to the processing method defined in advance for a specific event , Contains operations through （ If you send a notice ） And conditions （ When to perform the operation ）.
• Alarm upgrade （escalation）： A custom scheme for sending alarms or executing remote commands , Such as how often to send an alarm , How many times has it been sent .
• medium （media）： A means or channel for sending notices , Such as Email,Jabber or SMS etc. .
• notice （motification）： Information about an event sent to a user through a selected medium .
• Remote command （remote command）： Predefined commands , It can be automatically executed when the monitored host is under certain conditions .
• Template （template）： It is used to quickly define the preset entry set of the monitored host , It usually includes item,trigger,graph,screen,application,low-level discovery rule, Templates can be directly linked to a single host .
• application （application）： A group of item Set
• web scene （web scenario）： Used to detect web One or more aspects of site availability HTTP request .
• front end （frontend）： zabbix Of web Interface .

6. zabbix The components of

Zabbix_server： Server daemons

• Zabbix The core program of monitoring , The main function is to communicate with Zabbix proxies and Agents Interact 、 Trigger calculation 、 Send alert notification ; And save the data in a centralized way . Responsible for receiving agent The core component of the report information sent , All configurations 、 Statistical data and operational data are organized by

Zabbix_agentd：agent Daemon

• Deployed on the monitoring object , Programs that can actively monitor local resources and Applications , Deployed on the monitored host , Collect local data and send it to Server End or Proxy End

Zabbix_proxy： proxy server , Optional

• A help Zabbix Server collecting data , Share responsibility Zabbix Server The load of the program , Optional components , Commonly used in distributed monitoring environment , agent Server Collect part of the monitored data and send it to Server End

Zabbix_database： The storage system ,MySQL/PGSQL

• Dedicated to storing all configuration information , And have zabbix Data collected

Zabbix_web：Web GUI

• zabbix Of GUI Interface , Usually with server Running on the same machine

Zabbix_get： Command line tools , Test direction agent The client initiates a data collection request

Zabbix_sender： Command line tools , Test direction server End send data

Zabbix_java_gateway：java gateway

7. zabbix There are two monitoring modes of

Zabbix_agent There are two modes of detection ： Active mode and passive mode

Passive mode , It's also the default Zabbix Monitoring mode , Passive mode is relative to passive mode proxy Speaking of .proxy Sending data actively is active mode ,proxy wait for server The request to send data again is passive mode .

Use zabbix The benefits of active mode ： You can monitor remote devices that are not accessible ; Monitor local network instability areas ; When monitoring tens of thousands of projects, using agents can effectively share zabbix server The pressure of the ; simplify zabbix Maintenance of distributed monitoring .

Passive mode ： from server towards agent Issue instructions to get data , namely agent Passively get the data and return it to server,server Periodically agent Ask for data , The biggest problem with this general model is that it will increase server The amount of work , In an environment of hundreds of servers server Can't get the latest data in time , But it's also the default way of working .

Take the initiative ：agent request server Get a list of active monitoring items , And actively submit the data to be detected in the monitoring items to server/proxy
passive ：server towards agent Request to get the data of the monitor item ,agent Return the data .

Passive mode ： The default is zabbix_agent Passive mode

Active mode and passive mode mainly stand on zabbix_agent In terms of identity

1. Passive mode （zabbix-server Polling detection zabbix-agent）
2. Active mode （zabbix-agent Report to zabbix-server）

zabbix Active mode and passive mode selection
1. When （Queue） There are a large number of delay monitoring items in the queue
2. When the monitoring host exceeds 300+ , Active mode is recommended

8. Zabbix and Prometheus The difference between

	Prometheus	Zabbix
Time of issue	2016	2012
development language	go	c+php
performance	Support ten thousand as a unit	The upper limit is about 10000 node
Community support	Relatively inferior zabbix, But the number is increasing	Widely applied , Support more mature , Any problems encountered can be found
Container support	Not only support swarm Native clusters , And support kubernetes Monitoring of container clusters , It is the best solution for container monitoring at present	Zabbix Appeared earlier , At that time, the container was not born , Naturally, the support for containers is also relatively poor
Enterprise use	Basically use kubernetes With the container enterprise ,Prometheus Is the best choice	In traditional monitoring system , Especially in server related monitoring , Take the absolute advantage of
Deployment difficulty	There's only one core Server Components , One command to start	Multiple systems , A variety of monitoring information collection methods