Sophisticated system monitoring Tools-netdata

Today, wandering around the internet, accidentally found a monitoring tool: Netdata, feel a surprise, monitoring tools can be so beautiful!

Simple to understand, this tool is relatively new, monitoring system operation status is very powerful, in addition to monitoring the CPU, network card, disk, memory, process and so on, and can even monitor the system's CPU interrupt, context switch and so on very deep system operation state.
The new version also supports the monitoring of commonly used Web services such as Nginx, Mysql, Mongodb, Postgres, Redis, memcached, Elasticsearch, HAproxy, Squid, PHP-FPM and other common components.

Netdata Installation

Netdata installation is also very simple, the installation of the basic and there is no configuration, installed well can be used.

Install dependent packages

[[email protected] ~]# yum install -y  zlib-devel libuuid-devel libmnl-dev gcc make git autoconf autoconf-archive autogen automake pkgconfig curl

Rely on download Netdata installation after installation to install

[[email protected] ~]# wget https://github.com/firehol/netdata/archive/master.zip

Unzip and install

[[email protected] ~]# unzip master.zip [[email protected] ~]# cd netdata-master/[[email protected] netdata-master]# ./netdata-installer.sh

Install a successful interface

Uninstall script generated: ./netdata-uninstaller.sh --- We are done! ---   ^  |.-.   .-.   .-.   .-.   .-.   .  netdata                          .-.   .-  |   ‘-‘   ‘-‘   ‘-‘   ‘-‘   ‘-‘   is installed and running now!  -‘   ‘-‘    +----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+--->  enjoy real-time performance and health monitoring...

After the installation is successful, it is started by default and does not require any configuration, it can be viewed via web page, netdata own web interface, and does not require host installation Web service to provide support. Its default listening port is 19999, directly in the browser to enter the ip+ port on it. Take a look at the effect

From the right side of the column can be seen to monitor the system indicators are very much, in addition to expand monitoring other application services, as long as the configuration of the services to be monitored can be automatically identified and added, Netdata Monitoring Application service configuration in/etc/netdata/ PYTHON.D, the configuration of many services is supported by default, then configure an nginx monitor to see
Monitor Nginx must be open nginx Status monitoring page, here Skip, test nginx status page after normal access, you can configure/etc/netdata/python.d/ The nginx.conf file is added to the file or modified on the basis of the principle configuration:

localhost:  name : ‘local‘  url  : ‘http://192.168.214.128/status‘

Just open it. Nginx log monitoring, configuration is very simple, edit web_log.conf, add or modify the path of Nginx log

# centosnginx_log:  name: ‘nginx‘  path: ‘/var/log/nginx/access.log‘

And then restart the netdata to automatically identify the monitor page.

[[email protected] ~]# systemctl restart netdata

Now let's refresh the monitoring page.

So nginx monitoring completed, is not very simple, this tool on the single-machine monitoring is very comprehensive, but the cluster and distributed monitoring is not so simple, I hope to continue to improve the function bar. Second, it suddenly turns out that watching these monitors can make a clearer sense of how the system works.
Simple and practical to write so much, then have time to study slowly.

Sophisticated system monitoring Tools-netdata