YAML · 1114 bytes Raw Blame History
1 # Alertmanager — routes pages to the operator's pager and ticket-
2 # severity alerts to email. Webhook URL is templated from the
3 # operator's secret store; never commit a real one here.
4
5 global:
6 resolve_timeout: 5m
7
8 route:
9 receiver: tickets
10 group_by: [alertname, service]
11 group_wait: 30s
12 group_interval: 5m
13 repeat_interval: 6h
14 routes:
15 - matchers: [severity="page"]
16 receiver: pager
17 group_wait: 10s
18 repeat_interval: 1h
19
20 receivers:
21 - name: tickets
22 email_configs:
23 - to: ops@shithub.example
24 from: alertmanager@shithub.example
25 smarthost: smtp.shithub.example:587
26 auth_username: alertmanager@shithub.example
27 auth_password_file: /etc/alertmanager/smtp.password
28 require_tls: true
29 send_resolved: true
30
31 - name: pager
32 webhook_configs:
33 - url_file: /etc/alertmanager/pager.url
34 send_resolved: true
35
36 inhibit_rules:
37 # If the whole web tier is down, suppress per-route latency noise.
38 - source_matchers: [alertname="ShithubdWebDown"]
39 target_matchers: [alertname="HighRequestLatencyP95"]
40 equal: [cluster]
41