From 89314545911651d7f31e414fd1f86efebe4d87e7 Mon Sep 17 00:00:00 2001
From: "lin.huang" <lin.huang@apulis.com>
Date: Fri, 24 Nov 2023 16:06:42 +0800
Subject: [PATCH] feature:add redis error alert rule

---
 config/instance-down-rules.yml | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/config/instance-down-rules.yml b/config/instance-down-rules.yml
index 72b1f5aa3..719f673c6 100644
--- a/config/instance-down-rules.yml
+++ b/config/instance-down-rules.yml
@@ -8,4 +8,15 @@ groups:
           severity: critical
         annotations:
           summary: "Instance {{ $labels.instance }} down"
-          description: "{{ $labels.instance }} of job {{ $labels.job }} has been down for more than 1 minutes."
\ No newline at end of file
+          description: "{{ $labels.instance }} of job {{ $labels.job }} has been down for more than 1 minutes."
+
+  - name: database_insert_failure_alerts
+    rules:
+      - alert: DatabaseInsertFailed
+        expr: (increase(MsgInsertRedisFailedCounter[5m]) > 0) or (increase(MsgInsertMongoFailedCounter[5m]) > 0)
+        for: 1m
+        labels:
+          severity: critical
+        annotations:
+          summary: "Increase in MsgInsertRedisFailedCounter or MsgInsertMongoFailedCounter detected"
+          description: "Either MsgInsertRedisFailedCounter or MsgInsertMongoFailedCounter has increased in the last 5 minutes, indicating failures in message insert operations to Redis or MongoDB,maybe the redis or mongodb is crash."