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
@@ -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
|
||||
|
||||
Reference in new issue
Block a user