fix lokicompactorfailedcompaction rule (#1381)

Co-authored-by: Herve Nicol <12008875+hervenicol@users.noreply.github.com>
giantswarm · Sep 26, 2024 · e6e8d5f · e6e8d5f
1 parent b4bcd9b
commit e6e8d5f
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 1 deletion.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -7,6 +7,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Fixed
+
+- fix `LokiFailedCompaction` to take latest successfull compaction across multiple compactor/backend pods
+
 ## [4.16.0] - 2024-09-26
 
 ### Added

diff --git a/helm/prometheus-rules/templates/platform/atlas/alerting-rules/loki.rules.yml b/helm/prometheus-rules/templates/platform/atlas/alerting-rules/loki.rules.yml
@@ -143,7 +143,7 @@ spec:
         description: 'Loki compactor has been failing compactions for more than 2 hours since last compaction.'
         opsrecipe: loki#lokicompactorfailedcompaction
       # This alert checks if Loki's the last successful compaction run is older than 2 hours
-      expr: (time() - (loki_boltdb_shipper_compact_tables_operation_last_successful_run_timestamp_seconds > 0) > 60 * 60 * 2)
+      expr: (min by (cluster_id, installation, provider, pipeline) (time() - (loki_boltdb_shipper_compact_tables_operation_last_successful_run_timestamp_seconds > 0)) > 60 * 60 * 2)
       for: 1h
       labels:
         area: platform