18th Day of Python learning

The content of this section

Why do we need to monitor?

Discussion on the design of common monitoring system

Monitoring System Architecture Design

Structure Design of monitoring table

Why do we need to monitor? – Familiar with the design principle of it monitoring system – developing a simple version of the class Zabbix monitoring system – Mastering the program design ideas of the automation development project and the architecture of the common Monitoring system design discussion on Zabbixnagios Monitoring system requirements Discussion 1. Can monitor common system services, applications, Network equipment, etc. 2. A host can monitor a number of different services, different service monitoring interval can be different 3. The same service on different host monitoring interval, alarm threshold value can be different 4. You can add, delete, and modify the services to be monitored by a batch of hosts 5. Alarm level:

• Different services because of the degree of business importance different, if there is a problem can be set different alarm level
• You can specify a specific service or alert level event notification to a specific user
• Alarm Upgrade Settings

6. Storage and optimization of historical data

• Achieve the most efficient data storage with the least amount of space consumed
• How do I get 5 years of monitoring data from all the services on a single host within 1s?

7. Data visualization, how to make a simple and beautiful user interface?

8. How can I support 5000+ machine monitoring requirements for standalone machines? 9. What mode of communication is adopted? Active, passive?  10. How can I achieve horizontal scaling of the monitoring server? What architecture is used?? Mysql? Active communication? Snmp,wget ...? Passive communication? Agent---How do I communicate with the monitor server? Socket server–> sockect Client? Can I use a ready-made C/s architecture? Rabbit MQ, Redis subscription publishing, HTTP? Using HTTP Benefits

1. Simple interface Design

2. Easy horizontal scaling to do distributed

3.Socket stable and mature, eliminating the more communication maintenance energy

HTTP Features:

1. Short connection

2. No status

3. Safety Certification

4. Passive communication

Monitoring System Architecture Design

Table structure Design View Code

Classification:the path to Python's automated development

18th Day of Python learning