Merge branch 'main' into karpenter-alerts

giantswarm · Dec 9, 2024 · 7d2151f · 7d2151f
2 parents ed13e1e + b666430
commit 7d2151f
Show file tree

Hide file tree

Showing 3 changed files with 10 additions and 3 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -11,6 +11,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Add alerts for `karpenter` issues.
 
+## [4.29.0] - 2024-12-09
+
+### Changed
+
+- Increase time to trigger `PromtailRequestsErrors` alert from 15 to 25m.
+
 ## [4.28.0] - 2024-12-02
 
 ### Added
@@ -3303,7 +3309,8 @@ Fix `PromtailRequestsErrors` alerts as promtail retries after some backoff so ac
 
 - Add existing rules from https://github.com/giantswarm/prometheus-meta-operator/pull/637/commits/bc6a26759eb955de92b41ed5eb33fa37980660f2
 
-[Unreleased]: https://github.com/giantswarm/prometheus-rules/compare/v4.28.0...HEAD
+[Unreleased]: https://github.com/giantswarm/prometheus-rules/compare/v4.29.0...HEAD
+[4.29.0]: https://github.com/giantswarm/prometheus-rules/compare/v4.28.0...v4.29.0
 [4.28.0]: https://github.com/giantswarm/prometheus-rules/compare/v4.27.0...v4.28.0
 [4.27.0]: https://github.com/giantswarm/prometheus-rules/compare/v4.26.2...v4.27.0
 [4.26.2]: https://github.com/giantswarm/prometheus-rules/compare/v4.26.1...v4.26.2

diff --git a/helm/prometheus-rules/Chart.yaml b/helm/prometheus-rules/Chart.yaml
@@ -5,7 +5,7 @@ home: https://github.com/giantswarm/prometheus-rules
 icon: https://s.giantswarm.io/app-icons/1/png/default-app-light.png
 name: prometheus-rules
 appVersion: '0.1.0'
-version: '4.28.0'
+version: '4.29.0'
 annotations:
   application.giantswarm.io/team: "atlas"
   config.giantswarm.io/version: 1.x.x
diff --git a/helm/prometheus-rules/templates/platform/atlas/alerting-rules/promtail.rules.yml b/helm/prometheus-rules/templates/platform/atlas/alerting-rules/promtail.rules.yml
@@ -38,7 +38,7 @@ spec:
             opsrecipe: promtail/
           expr: |
             100 * (sum(rate(promtail_request_duration_seconds_count{status_code!~"2.."}[5m])) by (cluster_id, installation, provider, pipeline, namespace, job, route, instance) / sum(rate(promtail_request_duration_seconds_count[5m])) by (cluster_id, installation, provider, pipeline, namespace, job, route, instance)) > 10
-          for: 15m
+          for: 25m
           labels:
             area: platform
             severity: page