-
Bug
-
Resolution: Done
-
Normal
-
ACM 2.11.0
-
None
-
False
-
-
False
-
-
-
Search Sprint 2025-8
-
-
-
None
Description of problem:
On a 120 node managed cluster we are observing a high klusterlet-addon-search CPU usage and corresponding KAPI load:
# cat */audit* | jq -r '.user.username' | sort | uniq -c | sort -nr | head -n 10 854923 system:aggregator 483427 system:serviceaccount:open-cluster-management-agent-addon:klusterlet-addon-workmgr 335028 system:serviceaccount:openshift-monitoring:prometheus-k8s 324016 system:serviceaccount:openshift-sriov-network-operator:sriov-network-config-daemon 264290 system:serviceaccount:openshift-numaresources:rte 131442 system:serviceaccount:openshift-authentication-operator:authentication-operator 114368 system:serviceaccount:openshift-monitoring:cluster-monitoring-operator 103246 system:serviceaccount:openshift-console-operator:console-operator 93184 system:serviceaccount:openshift-cluster-version:default 92424 system:serviceaccount:openshift-controller-manager-operator:openshift-control # oc adm top pod -A | grep klusterlet open-cluster-management-agent klusterlet-6f7d64977c-h6tbt 5m 43Mi open-cluster-management-agent klusterlet-agent-69cbd5869c-bgxpw 14m 68Mi open-cluster-management-agent klusterlet-agent-69cbd5869c-k8r6z 2m 24Mi open-cluster-management-agent klusterlet-agent-69cbd5869c-tf2rr 2m 24Mi open-cluster-management-agent-addon klusterlet-addon-search-55fb47977c-zgd8f 1229m 228Mi open-cluster-management-agent-addon klusterlet-addon-workmgr-576644c785-qsj5f 207m 57Mi
Version-Release number of selected component (if applicable):
ACM: 2.11
Additional info:
- clones
-
-
- Closed
-