I have a Wordpress server, a PeerTube server, and a Prometheus server. I would like to have a single place to monitor the network i/o, RAM, CPU, and disk usage of all three servers (hence the Prometheus server).
I believe a single Prometheus server should be able to scrape metrics from multiple servers, provided each server has Node Exporter installed. I have installed Node Exporter on two of the servers but am confused as to whether I should configure Prometheus with one ‘job’ per server, e.g.
- job_name: 'node'
static_configs:
- targets: ['localhost:xxxx']
- job_name: 'node'
static_configs:
- targets: ['xxx.xxx.xxx.xxx:xxxx']
or configure Prometheus with only one job with one ‘target’ per server, e.g.:
- job_name: 'node'
static_configs:
- targets: ['localhost:xxxx', 'xxx.xxx.xxx.xxx:xxxx']
I’m using the node
job name because that’s what the Node Exporter - USE Method / Node dashboard seems to require, but I’m not married to that dashboard.
Thank you in advance for your help!