Operation and maintenance thinking | do you know the relationship between CMDB and monitoring?

Preface

With the development of business and new IT The constant introduction of Technology , Application system IT The scale of resources is growing ,IT The complexity of architecture is increasing . In this case , Need to go through a variety of monitoring systems , Different ways to sense whether the business system is alive , How are you living , How about the user experience . Common types of monitoring systems include ： Basic environmental monitoring 、 network monitoring 、 System monitoring 、 Database monitoring 、 Application monitoring 、 User experience monitoring and so on .

In this case , Our previous article 《 How to improve monitoring , Try to build a unified monitoring platform system for enterprises ！》 I have discussed , A unified monitoring center is needed to manage multiple alarm sources , In the middle of the alarm data processing , Provide monitoring data that can be consumed .

The overall architecture is as follows ：

Unified monitoring capability Architecture

There's a problem here , Monitoring and corporate CMDB What is the relationship between ？

---

CMDB And monitoring

Our understanding has the following two layers ：

• CMDB It is necessary to provide necessary supporting data for the monitoring system , Let's converge 、 Three dimensional 、 Standardized alarm information .
• CMDB We also need to get through to the monitoring system , Add... To new objects CMDB It can automatically add the object to the monitoring system ; At the same time, when the configuration data changes , It can send out necessary alarm information through the monitoring system .

Let's talk about the first level of relationship . The monitoring system , such as zabbix, When a certain monitoring indicator of an object reaches the threshold value , Will start warning ：XX Object's XX Indicator alarm and detailed information, etc . And you can zabbix Configure dependencies between monitoring items , Realize the convergence and correlation of alarms .

But there is a problem , Let's imagine a scenario ： You are a big 2C The company's DBA, Winter night, early morning 3 O 'clock , It's windy outside in the northwest , All of a sudden, the cell phone rings loudly , There is an alarm message to prompt the application system A Database nodes 01 Abnormal connection alarm . The content of warning information is limited , At this time, do you want to get up and turn on the computer for further processing ？

struggle , Right . Actually, as an administrator , When receiving this warning message , In addition to the need to know that there is a problem with this database , In fact, I want to know more about , such as ： What application system does this database belong to 、 In what environment 、 Is it a highly available cluster 、 Who is responsible for the application 、 Which application systems need to rely on this application system 、 Is there any configuration change in this database , So as to make further judgment and arrange the next operation ： For example, in the early morning of winter , Do you want to get up and turn on the computer . So at this point , We need a system that can provide ： Apply hierarchical topology 、 Cluster information 、 Module information 、 Resource instance 、 Information such as relationship , This system is CMDB.

CMDB Configuration Management

---

Integration and integration of the two

With CMDB after , When the alarm occurs , We can get the alarm system to CMDB Query the comprehensive configuration information related to this alarm object , In order to provide the most accurate 、 Rich and standard alarm information . for instance , In the last scene , If we know the database instance 01 It belongs to the application system A Of the test environment , And there are high availability clusters , Currently, there is no user accessing this database in theory , Why do administrators get up from the cold and turn on computers ？

On the contrary , If this database is found to be a system A Database of production environment , And because of the recent upgrade , Now it's single point mode , And then there's the system B and C Need to rely on the system A, Then hurry up and deal with the trouble , And notify B and C Launch the corresponding plan mechanism to minimize the impact .

here , Need CMDB Ability to provide data to monitoring system , Need to have corresponding data query 、 Interface information read , And it's easy to integrate .

Blue whale API gateway &CMDB Of API Interface list

On the other hand ,CMDB It also needs to synchronize its own data to the monitoring system . for instance , We have launched a batch of new virtual machine nodes of a certain system , Submit the completion sheet , Finish recording CMDB Configuration information , Do you manually configure it in the monitor again ？ Obviously not very reasonable , Right ？ This is the time CMDB Be able to actively push new object information to the monitoring system , The monitoring system is based on the existing monitoring template , Send out agent、 Configure monitoring protocol 、 Start monitoring, etc .

in addition , If CMDB Scan to find the actual configuration information of a host and the current CMDB The information stored in the library is inconsistent , It should also alarm through the monitoring system , Inform the administrator for further processing ？

So here you see , Monitoring system and CMDB There is a close connection between . What's more, there is often more than one monitoring system in an enterprise , If every monitoring system has to work with CMDB Do integration once , You must be tired to death . This requires the concept of monitoring middle station and unified alarm management , We don't need every monitoring system directly with CMDB Integrate , Only need to connect all monitoring systems to the unified alarm center module , From the unified alarm module to CMDB Monitoring docking , Share Information . such , When we send out every alarm , All can be based on CMDB Information in , Become standardized 、 Three dimensional alarm , Not a flat alarm . This kind of alarm can really highlight the value .

---

This article is edited and released by Tencent blue whale Zhiyun , Tencent blue whale Zhiyun （ Short for blue whale ） The software system is a set of systems based on PaaS Technology solutions for , Committed to building an industry-leading one-stop automatic operation and maintenance platform . At present, the community version has been launched 、 Enterprise Edition , Welcome to experience .

Official website ：https://bk.tencent.com/

Download link ：https://bk.tencent.com/download/

Community ：https://bk.tencent.com/s-mart/community/question