feat(monitoring): заменить Prometheus на vmalert
ansible-core-mon / deploy (push) Successful in 37s
ansible-core-net / deploy (push) Successful in 56s
ansible-core-secrets / deploy (push) Successful in 16s
ansible-ger-nod-02 / deploy (push) Successful in 1m22s
ansible-opnsense / deploy (push) Successful in 6s
ansible-rus-central / deploy (push) Successful in 39s
ansible-core-mon / deploy (push) Successful in 37s
ansible-core-net / deploy (push) Successful in 56s
ansible-core-secrets / deploy (push) Successful in 16s
ansible-ger-nod-02 / deploy (push) Successful in 1m22s
ansible-opnsense / deploy (push) Successful in 6s
ansible-rus-central / deploy (push) Successful in 39s
Алерты считаются по VictoriaMetrics и уходят в тот же Alertmanager, второй сбор метрик больше не нужен. Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
1 parent
28b8bcf708
commit
57aa12b9c5
10 files changed
+59
-79
No files matched your search
@@ -1,6 +1,6 @@
|
||||
---
|
||||
monitoring_dir: /opt/monitoring
|
||||
prometheus_retention: "30d"
|
||||
victoriametrics_retention: "30d"
|
||||
|
||||
# Цели node-exporter: все хосты из inventory, кроме исключений.
|
||||
# Отключить scrape на хосте: host_vars monitoring_scrape_enabled: false
|
||||
@@ -8,11 +8,14 @@ prometheus_retention: "30d"
|
||||
monitoring_scrape_exclude_hosts:
|
||||
- opnsense
|
||||
monitoring_node_exporter_port: 9100
|
||||
monitoring_scrape_job_name: node-exporter
|
||||
monitoring_env_label: prod
|
||||
|
||||
victoriametrics_image: victoriametrics/victoria-metrics:v1.153.0
|
||||
vmalert_image: victoriametrics/vmalert:v1.153.0
|
||||
vmagent_image: victoriametrics/vmagent:v1.153.0
|
||||
# Свой vmagent уже пишет node-exporter этих хостов. Повторный scrape даст двойные сэмплы.
|
||||
vmagent_central_scrape_skip:
|
||||
- core-mon
|
||||
- core-net
|
||||
cadvisor_image: gcr.io/cadvisor/cadvisor:v0.49.2
|
||||
xray_checker_image: kutovoys/xray-checker:v1.3.1
|
||||
xray_checker_subscription_url: "{{ lookup('env', 'XRAY_CHECKER_SUBSCRIPTION_URL', default='') }}"
|
||||
|
||||
@@ -2,7 +2,7 @@ groups:
|
||||
- name: homelab_alerts
|
||||
rules:
|
||||
- alert: HostDown
|
||||
expr: up == 0
|
||||
expr: up{job="integrations/node_exporter"} == 0
|
||||
for: 2m
|
||||
labels:
|
||||
severity: critical
|
||||
|
||||
@@ -8,8 +8,6 @@
|
||||
mode: "{{ item.mode }}"
|
||||
loop:
|
||||
- { path: "{{ monitoring_dir }}", owner: "root", group: "root", mode: "0755" }
|
||||
- { path: "{{ monitoring_dir }}/prometheus", owner: "root", group: "root", mode: "0755" }
|
||||
- { path: "{{ monitoring_dir }}/prometheus_data", owner: "65534", group: "65534", mode: "0777" }
|
||||
- { path: "{{ monitoring_dir }}/alertmanager", owner: "root", group: "root", mode: "0755" }
|
||||
- { path: "{{ monitoring_dir }}/alertmanager_data", owner: "65534", group: "65534", mode: "0777" }
|
||||
- { path: "{{ monitoring_dir }}/grafana", owner: "root", group: "root", mode: "0755" }
|
||||
@@ -22,6 +20,7 @@
|
||||
- { path: "{{ monitoring_dir }}/victoria-metrics-data", owner: "65534", group: "65534", mode: "0775" }
|
||||
- { path: "{{ monitoring_dir }}/vmagent", owner: "root", group: "root", mode: "0755" }
|
||||
- { path: "{{ monitoring_dir }}/vmagent/conf.d", owner: "root", group: "root", mode: "0755" }
|
||||
- { path: "{{ monitoring_dir }}/vmalert", owner: "root", group: "root", mode: "0755" }
|
||||
|
||||
- name: Сбросить список целей scrape node-exporter
|
||||
ansible.builtin.set_fact:
|
||||
@@ -35,13 +34,12 @@
|
||||
name: "{{ item }}"
|
||||
address: "{{ hostvars[item].monitoring_scrape_address | default(hostvars[item].ansible_host) }}"
|
||||
port: "{{ hostvars[item].monitoring_node_exporter_port | default(monitoring_node_exporter_port) }}"
|
||||
env: "{{ hostvars[item].monitoring_env | default(monitoring_env_label) }}"
|
||||
loop: "{{ groups['all'] | sort }}"
|
||||
when:
|
||||
- item not in monitoring_scrape_exclude_hosts
|
||||
- hostvars[item].monitoring_scrape_enabled | default(true) | bool
|
||||
|
||||
- name: Проверить, что в Prometheus есть хотя бы одна цель node-exporter
|
||||
- name: Проверить, что есть хотя бы одна цель node-exporter
|
||||
ansible.builtin.assert:
|
||||
that:
|
||||
- monitoring_scrape_targets | length > 0
|
||||
@@ -51,23 +49,13 @@
|
||||
ansible.builtin.debug:
|
||||
msg: "{{ monitoring_scrape_targets | map(attribute='name') | list | sort }}"
|
||||
|
||||
- name: Установить конфигурацию Prometheus
|
||||
ansible.builtin.template:
|
||||
src: prometheus.yml.j2
|
||||
dest: "{{ monitoring_dir }}/prometheus/prometheus.yml"
|
||||
owner: root
|
||||
group: root
|
||||
mode: "0644"
|
||||
register: prometheus_config
|
||||
|
||||
- name: Установить правила алертов Prometheus
|
||||
- name: Установить правила алертов vmalert
|
||||
ansible.builtin.copy:
|
||||
src: alerts.yml
|
||||
dest: "{{ monitoring_dir }}/prometheus/alerts.yml"
|
||||
dest: "{{ monitoring_dir }}/vmalert/alerts.yml"
|
||||
owner: root
|
||||
group: root
|
||||
mode: "0644"
|
||||
register: prometheus_alerts
|
||||
|
||||
- name: Установить конфигурацию Alertmanager
|
||||
ansible.builtin.template:
|
||||
@@ -177,11 +165,6 @@
|
||||
when: xray_checker_env is changed
|
||||
changed_when: true
|
||||
|
||||
- name: Пересоздать Prometheus, чтобы подхватить конфиг
|
||||
ansible.builtin.command: docker compose -f {{ monitoring_dir }}/compose.yaml up -d --force-recreate --no-deps prometheus
|
||||
when: prometheus_config is changed or prometheus_alerts is changed or monitoring_compose is changed
|
||||
changed_when: true
|
||||
|
||||
- name: Дождаться готовности Ntfy
|
||||
ansible.builtin.uri:
|
||||
url: "http://127.0.0.1:80/v1/health"
|
||||
@@ -217,6 +200,15 @@
|
||||
delay: 2
|
||||
until: victoriametrics_health.status == 200
|
||||
|
||||
- name: Дождаться vmalert
|
||||
ansible.builtin.uri:
|
||||
url: "http://127.0.0.1:8880/health"
|
||||
status_code: 200
|
||||
register: vmalert_health
|
||||
retries: 20
|
||||
delay: 2
|
||||
until: vmalert_health.status == 200
|
||||
|
||||
- name: Перезапустить Grafana, чтобы подхватить datasource
|
||||
ansible.builtin.command: docker compose -f {{ monitoring_dir }}/compose.yaml restart grafana
|
||||
when: grafana_datasources is changed or grafana_dashboards is changed or grafana_dashboard_provider is changed
|
||||
|
||||
@@ -1,23 +1,6 @@
|
||||
name: monitoring
|
||||
|
||||
services:
|
||||
prometheus:
|
||||
image: prom/prometheus:v3.2.1
|
||||
container_name: monitoring-prometheus
|
||||
restart: unless-stopped
|
||||
command:
|
||||
- "--config.file=/etc/prometheus/prometheus.yml"
|
||||
- "--storage.tsdb.path=/prometheus"
|
||||
- "--storage.tsdb.retention.time={{ prometheus_retention }}"
|
||||
- "--web.enable-lifecycle"
|
||||
volumes:
|
||||
- ./prometheus:/etc/prometheus:ro
|
||||
- ./prometheus_data:/prometheus
|
||||
ports:
|
||||
- "9090:9090"
|
||||
depends_on:
|
||||
- alertmanager
|
||||
|
||||
alertmanager:
|
||||
image: prom/alertmanager:v0.28.1
|
||||
container_name: monitoring-alertmanager
|
||||
@@ -76,7 +59,7 @@ services:
|
||||
ports:
|
||||
- "3000:3000"
|
||||
depends_on:
|
||||
- prometheus
|
||||
- vmsingle
|
||||
|
||||
vmsingle:
|
||||
image: {{ victoriametrics_image }}
|
||||
@@ -85,6 +68,7 @@ services:
|
||||
restart: unless-stopped
|
||||
command:
|
||||
- --storageDataPath=/victoria-metrics-data
|
||||
- --retentionPeriod={{ victoriametrics_retention }}
|
||||
volumes:
|
||||
- ./victoria-metrics-data:/victoria-metrics-data
|
||||
ports:
|
||||
@@ -138,4 +122,23 @@ services:
|
||||
- vmsingle
|
||||
- cadvisor
|
||||
|
||||
vmalert:
|
||||
image: {{ vmalert_image }}
|
||||
container_name: monitoring-vmalert
|
||||
restart: unless-stopped
|
||||
command:
|
||||
- -datasource.url=http://vmsingle:8428
|
||||
- -notifier.url=http://alertmanager:9093
|
||||
- -rule=/etc/vmalert/alerts.yml
|
||||
- -evaluationInterval=15s
|
||||
- -configCheckInterval=30s
|
||||
- -httpListenAddr=0.0.0.0:8880
|
||||
volumes:
|
||||
- ./vmalert:/etc/vmalert:ro
|
||||
ports:
|
||||
- "127.0.0.1:8880:8880"
|
||||
depends_on:
|
||||
- vmsingle
|
||||
- alertmanager
|
||||
|
||||
|
||||
@@ -5,8 +5,8 @@ datasources:
|
||||
uid: prometheus
|
||||
type: prometheus
|
||||
access: proxy
|
||||
url: http://prometheus:9090
|
||||
isDefault: true
|
||||
url: http://vmsingle:8428
|
||||
isDefault: false
|
||||
jsonData:
|
||||
timeInterval: 15s
|
||||
|
||||
@@ -15,7 +15,7 @@ datasources:
|
||||
type: prometheus
|
||||
access: proxy
|
||||
url: http://vmsingle:8428
|
||||
isDefault: false
|
||||
isDefault: true
|
||||
jsonData:
|
||||
timeInterval: 15s
|
||||
|
||||
@@ -1,27 +0,0 @@
|
||||
global:
|
||||
scrape_interval: 15s
|
||||
evaluation_interval: 15s
|
||||
|
||||
rule_files:
|
||||
- "/etc/prometheus/alerts.yml"
|
||||
|
||||
alerting:
|
||||
alertmanagers:
|
||||
- static_configs:
|
||||
- targets:
|
||||
- "alertmanager:9093"
|
||||
|
||||
scrape_configs:
|
||||
- job_name: "prometheus"
|
||||
static_configs:
|
||||
- targets: ["localhost:9090"]
|
||||
|
||||
- job_name: "{{ monitoring_scrape_job_name }}"
|
||||
static_configs:
|
||||
{% for target in monitoring_scrape_targets %}
|
||||
- targets:
|
||||
- "{{ target.address }}:{{ target.port }}"
|
||||
labels:
|
||||
host: "{{ target.name }}"
|
||||
env: "{{ target.env }}"
|
||||
{% endfor %}
|
||||
@@ -5,3 +5,10 @@
|
||||
labels:
|
||||
instance: "{{ inventory_hostname }}"
|
||||
cluster: "{{ vmagent_cluster }}"
|
||||
{% for target in monitoring_scrape_targets if target.name not in vmagent_central_scrape_skip %}
|
||||
- targets: ["{{ target.address }}:{{ target.port }}"]
|
||||
labels:
|
||||
instance: "{{ target.name }}"
|
||||
cluster: "{{ vmagent_cluster }}"
|
||||
host: "{{ target.name }}"
|
||||
{% endfor %}
|
||||
Reference in new issue
Block a user