Basic monitoring using bash, prometheus and Grafana
Let's monitor with Grafana, Prometheus, Pushgateway
Monitoring server architecture
Download Pushgateway and prometheus
https://prometheus.io/download/
Download Grafana
https://grafana.com/grafana/download
Pushgateway and prometheus services are uploaded, modify prometheus.yml
[root@monitoring ~]# netstat -ntpl | grep -i 909 tcp6 0 0 :::9090 :::* LISTEN 4528/./prometheus tcp6 0 0 :::9091 :::* LISTEN 4513/./pushgateway [root@monitoring ~]# [root@monitoring ~]# systemctl enable --now grafana-server
Create a bash script to collect metrics Your next task is to create a simple bash script that collects metrics like CPU usage and memory usage for individual processes.
Your script can be defined as a cron task that will run every second later.
To perform this task, you have multiple candidates.
You can run the main commands every second, analyze them with sed, and send the metrics to Pushgateway.
Shell to take resources, cpu, memory and Disk
[root@monitoring script]# sh recursos.sh 192.168.121.82 [root@monitoring script]#
We upload the metrics to pushgateway
[root@monitoring script]# sh load-cpu.sh
We see it reflected in the console
Automatically pushgateway metrics are also stored in the prometheus database
The data can now be read by grafana
Example
