From eacee20bf2ee23e2c271848adc005afd6d838e23 Mon Sep 17 00:00:00 2001
From: Grant Slater <github@firefishy.com>
Date: Fri, 3 Oct 2025 13:46:22 +0100
Subject: [PATCH] prometheus: Add alertmanager rules for S3 replication lag and
 failures

---
 .../templates/default/alert_rules.yml.erb     | 26 +++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/cookbooks/prometheus/templates/default/alert_rules.yml.erb b/cookbooks/prometheus/templates/default/alert_rules.yml.erb
index b770dba49..2d58610e6 100644
--- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb
+++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb
@@ -927,3 +927,29 @@ groups:
           alertgroup: web
         annotations:
           job_processing_rate: "{{ $value | humanizePercentage }}"
+  - name: aws
+    rules:
+      - alert: aws s3 replication lag
+        expr: aws_s3_replication_latency_maximum > 7200
+        for: 1h
+        keep_firing_for: 30m
+        labels:
+          alertgroup: aws
+        annotations:
+          s3_object_replication_lag: "{{ $value | humanizeDuration }}"
+      - alert: aws s3 replication failures
+        expr: aws_s3_operations_failed_replication_sum > 0
+        for: 1h
+        keep_firing_for: 30m
+        labels:
+          alertgroup: aws
+        annotations:
+          s3_object_replication_failures: "{{ $value }} objects"
+      - alert: aws s3 replication pending
+        expr: aws_s3_operations_pending_replication_maximum > 1000
+        for: 1h
+        keep_firing_for: 30m
+        labels:
+          alertgroup: aws
+        annotations:
+          s3_object_replication_pending: "{{ $value }} objects"
-- 
2.39.5