hosts

@@ -10,6 +10,8 @@
 
 viviane.adm.auro.re
 
+[aurore_server]
+
 [aurore_pve]
 merlin.adm.auro.re
 
@@ -41,21 +43,20 @@ litl.adm.auro.re
 log.adm.auro.re
 
 [aurore_testing_vm]
-pendragon.adm.auro.re
 
 ###############################################################################
 # OVH
 
+[ovh_server]
+
 [ovh_pve]
 horus.adm.auro.re
 
 [ovh_container]
 synapse.adm.auro.re
-phabricator.adm.auro.re
+wikijs.adm.auro.re
-wiki.adm.auro.re
 www.adm.auro.re
 proxy-ovh.adm.auro.re
-matrix-services.adm.auro.re
 
 [ovh_vm]
 serge.adm.auro.re
@@ -192,6 +193,8 @@ fl-4-2.borne.auro.re
 ###############################################################################
 # Pacaterie
 
+[pacaterie_server]
+
 [pacaterie_pve]
 mordred.adm.auro.re
 titan.adm.auro.re
@@ -270,6 +273,7 @@ ee-2-1.borne.auro.re
 ee-2-2.borne.auro.re
 eo-0-1.borne.auro.re
 eo-2-1.borne.auro.re
+eo-3-1.borne.auro.re
 ep-0-1.borne.auro.re
 ep-1-1.borne.auro.re
 ep-1-2.borne.auro.re
@@ -279,6 +283,8 @@ ep-1-3.borne.auro.re
 ###############################################################################
 # George Sand
 
+[gs_server]
+
 [gs_pve]
 lancelot.adm.auro.re
 odin.adm.auro.re
@@ -323,7 +329,6 @@ gd-1-2.borne.auro.re
 gd-2-1.borne.auro.re
 gd-3-1.borne.auro.re
 gd-4-1.borne.auro.re
-gd-4-2.borne.auro.re
 gd-5-1.borne.auro.re
 gd-5-2.borne.auro.re
 gd-garage-1.borne.auro.re
@@ -340,7 +345,6 @@ gf-0-1.borne.auro.re
 gf-1-1.borne.auro.re
 gf-2-1.borne.auro.re
 gf-3-1.borne.auro.re
-gf-3-2.borne.auro.re
 gf-4-1.borne.auro.re
 gf-5-1.borne.auro.re
 gg-5-1.borne.auro.re
@@ -349,6 +353,9 @@ gh-1-2.borne.auro.re
 
 ###############################################################################
 # Les Rives
+
+[rives_server]
+
 [rives_pve]
 thor.adm.auro.re
 loki.adm.auro.re
@@ -367,6 +374,8 @@ radius-rives.adm.auro.re
 routeur-rives.adm.auro.re
 
 [rives_unifi]
+r1-0-1.borne.auro.re
+r1-0-2.borne.auro.re
 r1-1-1.borne.auro.re
 r1-1-2.borne.auro.re
 r1-1-3.borne.auro.re
@@ -383,6 +392,7 @@ r1-3-3.borne.auro.re
 r1-3-4.borne.auro.re
 r1-3-5.borne.auro.re
 r1-3-6.borne.auro.re
+r2-0-1.borne.auro.re
 r2-1-1.borne.auro.re
 r2-1-2.borne.auro.re
 r2-1-3.borne.auro.re
@@ -430,11 +440,14 @@ r3-4-8.borne.auro.re
 
 # -aurore services
 [aurore:children]
+aurore_server
+aurore_pve
 aurore_vm
 
 
 # everything at ovh
 [ovh:children]
+ovh_server
 ovh_pve
 ovh_container
 ovh_vm
@@ -448,6 +461,7 @@ fleming_unifi
 
 # everything at pacaterie
 [pacaterie:children]
+pacaterie_server
 pacaterie_pve
 pacaterie_vm
 pacaterie_unifi
@@ -461,12 +475,14 @@ edc_unifi
 
 # everything at georgesand
 [gs:children]
+gs_server
 gs_pve
 gs_vm
 gs_unifi
 
 # everything at Les Rives
 [rives:children]
+rives_server
 rives_pve
 rives_vm
 rives_unifi
@@ -489,6 +505,11 @@ rives_vm
 
 # every server
 [server:children]
+ovh_server
+aurore_server
+pacaterie_server
+gs_server
+rives_server
 fleming_server
 edc_server
 

monitoring.yml

@@ -14,6 +14,9 @@
           {{ groups['fleming_pve'] + groups['fleming_vm'] | list | sort }}
     prometheus_unifi_snmp_targets:
       - targets: "{{ groups['fleming_unifi'] | list | sort }}"
+    prometheus_ipmi_targets:
+      - targets: |
+          {{ groups['fleming_pve'] + groups['fleming_server'] | list | sort }}
   roles:
     - prometheus
 
@@ -34,6 +37,9 @@
     prometheus_ups_snmp_targets:
       - ups-pn-1.ups.auro.re
       - ups-ps-1.ups.auro.re
+    prometheus_ipmi_targets:
+      - targets: |
+          {{ groups['pacaterie_pve'] + groups['pacaterie_server'] | list | sort }}
   roles:
     - prometheus
 
@@ -54,6 +60,9 @@
     prometheus_targets:
       - targets: |
           {{ groups['edc_pve'] + groups['edc_vm'] + groups['edc_server'] | list | sort }}
+    prometheus_ipmi_targets:
+      - targets: |
+          {{ groups['edc_pve'] + groups['edc_server'] | list | sort }}
     prometheus_unifi_snmp_targets:
       - targets: "{{ groups['edc_unifi'] | list | sort }}"
   roles:
@@ -77,6 +86,9 @@
       - ups-gk-1.ups.auro.re
     prometheus_pdu_snmp_targets:
       - pdu-ga-1.ups.auro.re
+    prometheus_ipmi_targets:
+      - targets: |
+          {{ groups['gs_pve'] + groups['gs_server'] | list | sort }}
   roles:
     - prometheus
 
@@ -98,6 +110,9 @@
           {{ groups['rives_pve'] + groups['rives_vm'] | list | sort }}
     prometheus_unifi_snmp_targets:
       - targets: "{{ groups['rives_unifi'] | list | sort }}"
+    prometheus_ipmi_targets:
+      - targets: |
+          {{ groups['rives_pve'] + groups['rives_server'] | list | sort }}
   roles:
     - prometheus
 
@@ -113,6 +128,9 @@
     prometheus_targets:
       - targets: |
           {{ groups['aurore_pve'] + groups['aurore_vm'] | list | sort }}
+    prometheus_ipmi_targets:
+      - targets: |
+          {{ groups['aurore_pve'] + groups['aurore_server'] | list | sort }}
     prometheus_postgres_targets:
       - targets: |
           {{ groups['bdd'] + groups['radius'] | list | sort }}
@@ -146,7 +164,7 @@
     # Prometheus targets.json
     prometheus_targets:
       - targets: |
-          {{ groups['ovh_pve'] + groups['ovh_vm'] | list | sort }}
+          {{ groups['ovh_pve'] + groups['ovh_vm'] + groups['ovh_container'] | list | sort }}
     prometheus_postgres_targets:
       - targets:
           - bdd-ovh.adm.auro.re
@@ -180,8 +198,13 @@
   roles:
     - prometheus_postgres
 
+# IPMI Exporters
+- hosts: server,pve
+  roles:
+    - prometheus_ipmi
+
 
 # Monitor all hosts
-- hosts: all,!edc_unifi,!fleming_unifi,!pacaterie_unifi,!gs_unifi,!rives_unifi,!aurore_testing_vm,!ovh_container
+- hosts: all,!edc_unifi,!fleming_unifi,!pacaterie_unifi,!gs_unifi,!rives_unifi,!aurore_testing_vm
   roles:
     - prometheus_node

roles/prometheus/tasks/main.yml

@@ -113,6 +113,13 @@
     mode: 0644
   when: prometheus_pdu_snmp_targets is defined
 
+- name: Configure Prometheus ipmi targets monitoring
+  copy:
+    content: "{{ prometheus_ipmi_targets | to_nice_json }}\n"
+    dest: /etc/prometheus/targets_ipmi.json
+    mode: 0644
+  when: prometheus_ipmi_targets is defined
+
 - name: Activate prometheus service
   systemd:
     name: prometheus

roles/prometheus/templates/postgres.rules.yml.j2

@@ -112,7 +112,7 @@ groups:
             rate(pg_stat_database_xact_rollback{datname!~"template.*"}[3m]) /
             rate(pg_stat_database_xact_commit{datname!~"template.*"}[3m])
           ) * 100 
-          > 7
+          > 20
         for: 0m
         labels:
           severity: warning

roles/prometheus/templates/prometheus.yml.j2

@@ -140,4 +140,22 @@ scrape_configs:
       - target_label: __address__
         replacement: 127.0.0.1:9116
 
+  - job_name: ipmi
+    file_sd_configs:
+      - files:
+        - '/etc/prometheus/targets_ipmi.json'
+    metrics_path: /metrics
+    params:
+      module: [default]
+    relabel_configs:
+      # Do not put :9290 in instance name, rather here
+      - source_labels: [__address__]
+        target_label: __param_target
+      - source_labels: [__param_target]
+        target_label: instance
+      - source_labels: [__param_target]
+        target_label: __address__
+        replacement: '$1:9290'
+
+
 ...

roles/prometheus/templates/server.rules.yml.j2

@@ -11,7 +11,7 @@ groups:
     rules:
 
       - alert: InstanceDown
-        expr: up{instance!~".*.borne.auro.re$"} == 0
+        expr: up{instance!~".*.borne.auro.re$", job="servers"} == 0
         for: 3m
         labels:
           severity: critical
@@ -50,7 +50,7 @@ groups:
               node_memory_SwapFree_bytes
               / node_memory_SwapTotal_bytes
             )
-          ) * 100 >= 20
+          ) * 100 >= 50
         for: 3m
         labels:
           severity: warning

roles/prometheus_federate/templates/prometheus.yml.j2

@@ -39,4 +39,5 @@ scrape_configs:
         - '{job="apc_pdu_snmp"}'
         - '{job="docker"}'
         - '{job="switch_snmp"}'
+        - '{job="ipmi"}'
 ...

roles/prometheus_ipmi/files/ipmi_conf/ipmi.sh

@@ -0,0 +1,3 @@
+#!/bin/sh
+sudo /usr/sbin/$(basename $0) "$@"
+

roles/prometheus_ipmi/files/prometheus

@@ -0,0 +1,9 @@
+# Prometheus can be sudo for ipmi collector
+
+prometheus ALL = NOPASSWD: /usr/sbin/ipmimonitoring,\
+                              /usr/sbin/ipmi-sensors,\
+                              /usr/sbin/ipmi-dcmi,\
+                              /usr/sbin/ipmi-raw,\
+                              /usr/sbin/bmc-info,\
+                              /usr/sbin/ipmi-chassis,\
+                              /usr/sbin/ipmi-sel

roles/prometheus_ipmi/handlers/main.yml

@@ -0,0 +1,6 @@
+---
+- name: Restart prometheus-ipmi-exporter
+  service:
+    name: prometheus-ipmi-exporter
+    state: restarted
+    daemon_reload: true

roles/prometheus_ipmi/tasks/main.yml

@@ -0,0 +1,65 @@
+---
+- name: Install Prometheus ipmi-exporter
+  apt:
+    update_cache: true
+    name: prometheus-ipmi-exporter
+  register: apt_result
+  retries: 3
+  until: apt_result is succeeded
+
+- name: Make Prometheus ipmi-exporter scrap on localhost only
+  lineinfile:
+    path: /etc/default/prometheus-ipmi-exporter
+    regexp: '^ARGS='
+    line: |
+      ARGS="--config.file=/etc/prometheus/ipmi_local.yml  --freeipmi.path='/etc/prometheus/ipmi_conf/'"
+  notify: Restart prometheus-ipmi-exporter
+
+- name: Define prometheus to be sudo for some command
+  copy:
+    src: files/prometheus
+    dest: /etc/sudoers.d/prometheus
+    owner: root
+    group: root
+    mode: u=r,g=r,o=
+  notify: Restart prometheus-ipmi-exporter
+
+- name: Override commands launchs by prometheus for ipmi
+  copy:
+    src: files/ipmi_conf
+    dest: /etc/prometheus
+    owner: prometheus
+    group: prometheus
+    mode: u=rx,g=r,o=
+  notify: Restart prometheus-ipmi-exporter
+
+- name: Override commands launchs by prometheus for ipmi
+  file:
+    src: '/etc/prometheus/ipmi_conf/ipmi.sh'
+    dest: '/etc/prometheus/ipmi_conf/{{ item }}'
+    state: link
+    force: true
+    owner: prometheus
+    group: prometheus
+    mode: u=rx,g=,o=
+  loop:
+    - bmc-info
+    - ipmi-chassis
+    - ipmi-dcmi
+    - ipmimonitoring
+    - ipmi-raw
+    - ipmi-sensors
+    - ipmi-sel
+  notify: Restart prometheus-ipmi-exporter
+
+- name: Activate prometheus-ipmi-exporter service
+  systemd:
+    name: prometheus-ipmi-exporter
+    enabled: true
+    state: started
+
+- name: Restart prometheus-ipmi-exporter
+  systemd:
+    state: restarted
+    daemon_reload: true
+    name: prometheus-ipmi-exporter