Collect data on interrupts too

playbooks/prod_db.yml

@@ -3,6 +3,18 @@
 # Database servers
 ---
 - hosts: vm-general-1.ashburn.mgmt.desu.ltd
+  tasks:
+    - name: assure prometheus psql exporter
+      ansible.builtin.docker_container:
+        name: prometheus-psql-exporter
+        image: quay.io/prometheuscommunity/postgres-exporter
+        env:
+          DATA_SOURCE_URI: "10.0.0.2:5432/postgres"
+          DATA_SOURCE_USER: "nagios"
+          DATA_SOURCE_PASS: "{{ secret_postgresql_monitoring_password }}"
+        ports:
+          - 9102:9187/tcp
+      tags: [ db, psql, prometheus, monitoring, docker ]
   roles:
     - role: geerlingguy.postgresql
       vars:

playbooks/tags_nagios.yml

@@ -42,6 +42,8 @@
         image: quay.io/prometheus/node-exporter:latest
         command:
           - '--path.rootfs=/host'
+          - '--collector.interrupts'
+          - '--collector.processes'
         network_mode: host
         pid_mode: host
         volumes:

roles/prometheus/templates/prometheus.yml.j2

@@ -1,17 +1,17 @@
 # my global config
+# vim:ft=ansible:
 ---
 global:
   scrape_interval: 15s
   evaluation_interval: 15s
 
 scrape_configs:
-  # The job name is added as a label `job=<job_name>` to any timeseries
-  # scraped from this config.
+    # Default Prometheus job to monitor itself
   - job_name: "prometheus"
-    # metrics_path defaults to '/metrics'
-    # scheme defaults to 'http'.
     static_configs:
       - targets: ["localhost:9090"]
+
+    # These two jobs are included for every node in our inventory
   - job_name: "node-exporter"
     static_configs:
       - targets:
@@ -24,3 +24,11 @@ scrape_configs:
 {% for host in groups['tags_nagios'] %}
           - '{{ host }}:9101'
 {% endfor %}
+
+    # This container is used on every psql machine
+  - job_name: "psql-exporter"
+    static_configs:
+      - targets:
+{% for host in groups['tags_nagios-checkpgsql'] %}
+          - '{{ host }}:9102'
+{% endfor %}