The network related indicators of the node_exporter are inaccurate

General Help/Support
1

Hello everyone, I am new to Prometheus. I have tested node_exporter with Prometheus, and I am attempting to monitor network-related metrics such as net_conntrack_listener_conn_accepted_total . However, the graphs in Prometheus seem incorrect. Here is my configuration and the method I used.

prometheus configuration

global:
  scrape_interval:     5s 
  evaluation_interval: 5s 

alerting: 
  alertmanagers:
  - static_configs:
    - targets:
      
rule_files:

scrape_configs: 
  - job_name: 'linux-local' 
  	static_configs:
  	- targets: ['localhost:9100']

start prometheus

docker run -d -p 9090:9090 --restart=on-failure \
  -v /etc/prometheus/:/etc/prometheus/ \
  -v /etc/prometheus/prometheus.yml:/etc/prometheus/prometheus.yml \
  --name prometheus \
  prom/prometheus

Test shell

for ((i=1; i<=1000; i++)); do
  nc -zv xx.xx.xx.xx 22
done

image
image2380×916 106 KB

Thanks for your answers.

2

Hey Ryan, can you be more explicit in what you think it is incorrect?