当前位置:网站首页>[performance test] jmeter+grafana+influxdb deployment practice
[performance test] jmeter+grafana+influxdb deployment practice
2022-07-05 19:02:00 【bulabula2022】
Jmeter Configure the test results to be automatically sent to influxdb
There's a lot of information online ,jmeter Configuration is not described in detail here
If you want to Grafana More flexible display of data , Need to be right jmeter Plug in secondary development , Automatically store the test data into influxdb Time series database , stay grafana Panel customization sql Inquire about , Automatically display the desired data ,grafana The panel also needs secondary development according to the display needs
influxDB Deploy
1.linux Specify the directory to download InfluxDB
Download address :wget https://dl.influxdata.com/influxdb/releases/influxdb-1.8.4.x86_64.rpm
2. install InfluxDB:
yum localinstall influxdb-1.8.4.x86_64.rpm, Two main file locations

3. Change the service startup configuration :vim influxdb.service
1) take User and Group Value changed to root

2)influxdb.service file Restart the effective service command :systemctl daemon-reload
4. modify influxDB The configuration file
Carry out orders :vi /etc/influxdb/influxdb.conf, The final modified configuration is as follows :
[[graphite]]
# Determines whether the graphite endpoint is enabled.
enabled = true
database = "jmeter" # Database name
retention-policy = ""
bind-address = ":2003" # port
protocol = "tcp"
consistency-level = "one"
……
[http]
# Determines whether HTTP endpoint is enabled.
enabled = true
# The bind address used by the HTTP service.
bind-address = ":8086"
Execute after modification :wq Save and exit
5. start-up Influxdb service
1) Temporary start up :systemctl start influxdb
2) Boot up :systemctl enable influxdb
3) View running status :ps -ef | grep -i "influx" perhaps service influxdb status

4) restart :service influxdb restart
influxDB Of tcp The default listening port is 8088, perform :netstat -antp|grep 8088, The following display indicates that the port is restarted successfully

attach influxdb Some port descriptions
start-up influxdb database :
service influxdb start
stop it influxdb database :
service influxdb stop
Port specification
8083: visit web Address of the page ,8083 Is the default port ;1.4 Version this feature is deprecated , Change it to Chronograf
8086: Data writing influxdb The address of ,8086 Is the default port ;
8088: Data backup and recovery address ,8088 Is the default port ;
If linux Turn on the firewall , Need to open 8086 port , The order is as follows :
firewall-cmd --add-port=8086/tcp --permanent
firewall-cmd --add-port=8086/udp --permanent
firewall-cmd --reload
6. Create databases and users
1) Get into influx command :influx

2) Display database , and mysql similar :show databases

3) Create database : CREATE DATABASE jmeter( Library name )
> CREATE DATABASE jmeter
> use jmeter
> create user "admin" with password ' admin ' with all privileges
The user name and password set here need to be recorded , Later configuration JMeter and Grafana When it comes to
notes : Pay attention here ,CREATE DATABASE You have to use uppercase
Port specification :
2003 port :Jmeter The port to send data to the database , The server side also receives data through this port , Therefore, the server needs to open this 2003 port
8086 port ,Grafana or Chronograf from influxdb The port on which the database fetches data
4) Switch database :use jmeter( Library name ); Show me the table :show measurements

Grafana Deploy
Grafana It's one for Go Language development of open source data visualization tools , Can do data monitoring and data statistics , With alarm function
windows Version installation
1、 Official website download address :Download Grafana | Grafana Labs

2、 Unzip it to the installation folder
3、 start-up , Switch to Grafana In the installation directory bin Under the table of contents , double-click grafana-server.exe Start the program
4、 visit http://localhost:3000, Use admin/admin Sign in Grafana that will do , Here's the picture

Linux Version installation
1. Switch to the specified directory to download :wget https://dl.grafana.com/oss/release/grafana-7.4.3-1.x86_64.rpm
2. install :sudo yum install grafana-7.4.3-1.x86_64.rpm The configuration file grafana.ini Location

3. start-up grafana
1) Boot up
Start the service :systemctl start grafana-server
Boot up :systemctl enable grafana-server
Restart :systemctl restart grafana-server
View startup status :service grafana-server status
grafana Integrate influxdb
visit http://localhost:3000, Use admin/admin Sign in
1. Add data sources :Configration –>Data Sources –> Add data source à choice InfluxDB



2. To configure influxdb database


3. preservation , Show Data source is working It means that the addition is successful

More items may need to be completed
1.InfluxDB+Grafana Monitor panel configuration
Monitor panel configuration And more monitoring template optimization , In the follow-up implementation process, a kind of supplement
2.Jmeter+InfluxDB+Grafana Integration of the three , Need secondary development optimization
3.jenkins+jmeter+influxdb Continuous integrated monitoring
4.Prometheus+Grafana Integrated monitoring server (linux,wondows)
5.Prometheus+Grafana Integrated monitoring database (mysql,PostgreSQL,oracle)
6.nmon2influxdb+grafana Visual deployment of service monitoring
Secondary optimization or development , Problems to be solved
Q1:Jmeter Load Test panel -Active Users Show the number of virtual users 1 individual ( The implemented ) |
Q2:Jmeter Load Test panel -request Add fuzzy search function ( Accumulate more and more requests , Easy to search for )( The implemented ) |
Q3:Jmeter Load Test panel - Increase throughput real-time display panel ( Convenient for gradient pressure measurement TPS Model trend view )( The implemented ) |
Q4:Jmeter Load Test panel -metrics overview The display under the panel shows the real-time average throughput ( Currently, the running time is selected manually to participate in the calculation )( The implemented ) |
Q5:Jmeter Load Test panel - Add monitoring panel for server resource consumption ( The implemented ) |
Q6:Jmeter Load Test panel - Add to mysql/oracle/pg Wait for database performance monitoring data collection ( The implemented ) |
Q7:Jmeter Load Test panel - Some overview data can be drilled , View specific data details |
Q8:Jmeter Load Test panel - Add the display of pressure measurement data according to transactions ( Be right jmeter Secondary development of plug-in tools ) |
Q9:Jmeter Load Test panel - Drop down filtering function is added ( It mainly adds variables for multi-dimensional query ) |
Q9:Jmeter Load Test panel -grafana Plot configuration query variables + Multi level variable linkage + Query results are configured in combination with application scenarios ( By project , Function details have not been realized yet ) Reference resources :https://clevercode.blog.csdn.net/article/details/104324039?utm_medium=distribute.pc_relevant.none-task-blog-2%7Edefault%7EBlogCommendFromMachineLearnPai2%7Edefault-4.base&depth_1- utm_source=distribute.pc_relevant.none-task-blog-2%7Edefault%7EBlogCommendFromMachineLearnPai2%7Edefault-4.base |
grafana Historical data query optimization () |
Grafana error Log information can be viewed in real time () |
边栏推荐
- 2022最新Android面试笔试,一个安卓程序员的面试心得
- A2L file parsing based on CAN bus (3)
- SAP feature description
- 图扑软件数字孪生智慧风电系统
- 常见时间复杂度
- Is it safe to make fund fixed investment on access letter?
- How to quickly advance automated testing? Listen to the personal feelings of the three bat test engineers
- 图扑软件数字孪生 | 基于 BIM 技术的可视化管理系统
- lombok @Builder注解
- 7-1 linked list is also simple fina
猜你喜欢

MySQL数据库索引教程(超详细)

How to automatically install pythn third-party libraries

Analysis of postman core functions - parameterization and test report

Oracle Chinese sorting Oracle Chinese field sorting

Redhat7.4 configure Yum software warehouse (rhel7.4)

公司破产后,黑石们来了

The worse the AI performance, the higher the bonus? Doctor of New York University offered a reward for the task of making the big model perform poorly

图扑软件数字孪生 | 基于 BIM 技术的可视化管理系统

Postman核心功能解析 —— 参数化和测试报告

跨境支付平台 XTransfer 的低代码实践:如何与其他中台融合是核心
随机推荐
Optimization of middle alignment of loading style of device player in easycvr electronic map
Oracle 中文排序 Oracle 中文字段排序
UDF implementation of Dameng database
Ant group open source trusted privacy computing framework "argot": open and universal
Shang Silicon Valley Shang preferred project tutorial release
ICML2022 | 长尾识别中分布外检测的部分和非对称对比学习
MySQL数据库索引教程(超详细)
AI Open2022|基于异质信息网络的推荐系统综述:概念,方法,应用与资源
Taishan Office Technology Lecture: from the layout height of the line, talk about the height overflow and height shrinkage of the drawing height (launched in the whole network)
为什么 BI 软件都搞不定关联分析?带你分析分析
深入底层C源码讲透Redis核心设计原理
The road of enterprise digital transformation starts from here
[HCIA cloud] [1] definition of cloud computing, what is cloud computing, architecture and technical description of cloud computing, Huawei cloud computing products, and description of Huawei memory DD
解决 contents have differences only in line separators
AI open2022 | overview of recommendation systems based on heterogeneous information networks: concepts, methods, applications and resources
Redhat7.4 configure Yum software warehouse (rhel7.4)
MySQL优化六个点的总结
AI表现越差,获得奖金越高?纽约大学博士拿出百万重金,悬赏让大模型表现差劲的任务
公司破产后,黑石们来了
R语言使用lubridate包处理日期和时间数据实战