title: “2. Prometheus installieren”
date: 2016-10-07T13:00:36
slug: 2-prometheus-installieren

wget https://github.com/prometheus/prometheus/releases/download/v1.1.3/prometheus-1.1.3.linux-amd64.tar.gz
tar -xzf prometheus-1.1.3.linux-amd64.tar.gz
cd prometheus-1.1.3.linux-amd64

vi prometheus.yml

global:
 scrape\_interval: 15s # By default, scrape targets every 15 seconds.
 evaluation\_interval: 15s # Evaluate rules every 15 seconds.

 # Attach these extra labels to all timeseries collected by this Prometheus instance.
 external\_labels:
 monitor: 'codelab-monitor'

rule\_files:
 - 'prometheus.rules'

scrape\_configs:
 - job\_name: 'prometheus'

 # Override the global default and scrape targets from this job every 5 seconds.
 scrape\_interval: 5s

 static\_configs:
 - targets: ['localhost:9090']

 - job\_name: 'node'

 # Override the global default and scrape targets from this job every 5 seconds.
 scrape\_interval: 1s

 static\_configs:

 - targets: ['127.0.0.1:9100']
 labels:
 group: 'localhost'

Diese Konfiguration pollt jede Sekunde die Metrics vom localhost port 9100

Einen Job anlegen (rules Datei):
vi prometheus.rules

job\_service:bandwidth1 = rate(node\_network\_receive\_bytes{device="eth0"}[2s])

Einen Alert anlegen (rules Datei):

# Bandwidth Alert > 1s.
ALERT BandwidthHighAlert
 IF rate(node\_network\_receive\_bytes{device="eth0"}[2s])/1024/1024 > 1
 FOR 10s
 ANNOTATIONS {
 summary = "high Network Traffic on {{ $labels.instance }}",
 description = "{{ $labels.instance }} has high Network Traffic (current value: {{ $value }}s)",
 }

Start Prometheus with AlertManager at localhost Port 9093

./prometheus -config.file=prometheus.yml -alertmanager.url=http://localhost:9093