add prometheus metrics for errors when getting certificates through acme (typically from let's encrypt)

and add an alerting rule for it. we certainly want a heads up when there are issues with the certificates.
2025-07-12 17:04:39 +03:00 · 2025-02-06 15:12:36 +01:00
parent 1277d78cb1
commit e5e15a3965
2 changed files with 29 additions and 2 deletions
--- a/prometheus.rules
+++ b/prometheus.rules
@ -8,6 +8,11 @@ groups:
    annotations:
      summary: unhandled panic

+  - alert: mox-acme-request-cert-errors
+    expr: increase(mox_autotls_cert_request_errors_total[1h]) > 0
+    annotations:
+      summary: errors requesting tls certificates with acme
+
  - alert: mox-ip-on-dns-blocklist
    expr: mox_dnsbl_ips_success < 1
    annotations: