Prometheus use of indicators to monitor Linux system

Mounting a first sensor node explorer, Download on Linux systems https://prometheus.io/docs/guides/node-exporter/

This detector will periodically index the hardware and kernel parameters linux system by 9100 and url metrics port exposed to the outside.

clipboard1,1

Start node explorer, is displayed on the monitor port 9100:

clipboard2,2

Browser, enter hostname: 9100 / metrics can access to the parameters of the node explorer collected:

clipboard3,3

Some examples:

CPU usage time under different operating modes: node_cpu_seconds_total

clipboard4,4

Number of bytes available file system: node_filesystem_avail_bytes

clipboard5,5

The number of bytes received from the network card: node_network_receive_bytes_total

clipboard6,6

Prometheus below with reference to collect and display exposed through node explorer data.

Prometheus is defined in prometheus.yml file in the installation directory of a job, pointing to node explorer running on a Linux system:

clipboard7,7

Local start Prometheus, open the following URL:
HTTP: // localhost: 9090 /
input node_cpu_seconds_total {mode = "system"} , the query time on all servers in the system working state of CPU consumption:

clipboard8,8

Can also specify a time window, only the CPU running the query data in the last 1 minute:
Rate (node_cpu_seconds_total {MODE = "System"} [1M])

clipboard9,9

For more Jerry's original article, please pay attention to the public number "Wang Zixi":
公众号截图

Guess you like

Origin www.cnblogs.com/sap-jerry/p/11954603.html