-
Bug
-
Resolution: Duplicate
-
Major
-
None
-
4.16
-
Important
-
No
-
Approved
-
False
-
-
Description of problem:
iptables-alerter pod under openshift-network-operator namespace is using up a large chunk of cpu on SNO with DU configuration Without workload (idle-state): Avg: 0.019 cores Max: 1.063 cores With reference DU workload: Avg: 0.020 cores Max: 1.022 cores Overall openshift-network-operator sees a ~380.5% in cpu usage between ocp 4.15.2 and 4.16.0-rc.1
Version-Release number of selected component (if applicable):
ocp 4.16.0-rc.1
How reproducible:
1. Install OCP 4.16.0-rc.1 SNO and apply DU configs 2. Observe cpu usage over a steady period of two hours each (6 iterations) with and without workloads
Steps to Reproduce:
1. 2. 3.
Actual results:
openshift-network-operator namespace cpu usage is 380% higher in 4.16.0-rc.1 compared to ocp 4.15
Expected results:
Cpu usage of openshift-network-operator namespace between releases should be comparable
Additional info: