Description of problem: Nodes with a high number of NICs cause high CPU load even if NICs are ignored by node_exporter. See also https://issues.redhat.com/browse/MON-1339 Version-Release number of selected component (if applicable): How reproducible: Start a single node cluster. Steps to Reproduce: 1. 2. 3. Actual results: CPU utilization is high, no matter how many NICs are ignored. Expected results: CPU utilization should decrease as NICs are ignored. Additional info:
tested with 4.9.0-0.nightly-2021-07-21-081948 single-node cluster # oc get no NAME STATUS ROLES AGE VERSION ip-10-0-151-142.us-east-2.compute.internal Ready master,worker 52m v1.21.1+8268f88 total cpu requests for node-exporter is 9Mi node-exporter-pv4f8 Container Name: node-exporter resources: map[requests:map[cpu:8m memory:32Mi]] Container Name: kube-rbac-proxy resources: map[requests:map[cpu:1m memory:15Mi]] watched for a while, cpu usage is about 9Mi, no sudden increase sum(kube_pod_resource_request{resource='cpu',pod='node-exporter-pv4f8',namespace='openshift-monitoring'}) BY (pod, namespace) {namespace="openshift-monitoring", pod="node-exporter-pv4f8"} 0.009000000000000001
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Container Platform 4.9.0 bug fix and security update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2021:3759