commit

jobs/templates/cert-checker.yml

@@ -0,0 +1,5 @@
+loglevel: debug
+port: 8080  # Optional
+intervalminutes: 2
+certificates:
+    - dns: auth.tservic.es

jobs/templates/node_exporter.service.j2

@@ -0,0 +1,10 @@
+[Unit]
+Description=Node Exporter
+After=network.target
+
+[Service]
+Type=simple
+ExecStart=/usr/local/bin/node_exporter --web.config.file="/etc/node_exporter/config.yml" --web.listen-address=":9100"
+
+[Install]
+WantedBy=multi-user.target

jobs/templates/prometheus.yml.j2

@@ -0,0 +1,56 @@
+---
+global:
+  scrape_interval: 15s
+  evaluation_interval: 15s
+
+scrape_configs:
+  - job_name: cert-checker
+    scrape_interval: 5s
+    static_configs:
+      - targets:
+        - cert-checker:8080
+
+  - job_name: "nodeexporter"
+    static_configs:
+      - targets: ["node-exporter:9100"]
+
+  - job_name: "cadvisor"
+    static_configs:
+      - targets: ["cadvisor:8080"]
+{% for host in groups['docker']%}
+{% if inventory_hostname != host %}
+      - targets: ["{{ host }}{{hostsuffix}}:9999"]
+{% endif %}
+{% endfor %}
+
+  - job_name: "prometheus"
+    static_configs:
+      - targets: ["localhost:9090"]
+
+  - job_name: "node_exporter"
+    static_configs:
+{% for host in groups['locations'] %}
+{% if inventory_hostname != host and host not in groups['windows'] %}
+      - targets: ["{{ host }}{{hostsuffix}}:9100"]
+{% endif %}
+{% endfor %}
+
+
+  - job_name: "docker"
+    static_configs:
+{% for host in groups['docker'] %}
+{% if inventory_hostname != host %}
+      - targets: ["{{ host }}{{hostsuffix}}:9323"]
+{% endif %}
+{% endfor %}
+
+
+rule_files:
+  - rules.yml
+
+alerting:
+  alertmanagers:
+    - static_configs:
+      - targets:
+        # Alertmanager's default port is 9093
+        - localhost:9093

jobs/templates/rules.yml

@@ -0,0 +1,14 @@
+groups:
+- name: AllInstances
+  rules:
+  - alert: InstanceDown
+    # Condition for alerting
+    expr: up == 0
+    for: 1m
+    # Annotation - additional informational labels to store more information
+    annotations:
+      title: 'Instance {{ $labels.instance }} down'
+      description: '{{ $labels.instance }} of job {{ $labels.job }} has been down for more than 1 minute.'
+    # Labels - additional labels to be attached to the alert
+    labels:
+      severity: 'critical'