Description of problem:
On an OpenShift 4.11.0-rc.1 configured with OVN-Kubernetes, we observed significant regression in the kube-apiserver CPU and Memory usage from 4.10 to 4.11.
When compared to Openshift SDN, the kube-apiserver resources grow at a significantly higher rate than an OpenShift SDN cluster installed with the same version.
Version-Release number of selected component (if applicable):
4.11.0-rc.1
How reproducible:
100%
Steps to Reproduce:
1. Install a 4.11.0-rc.1 cluster with r5.4xlarge masters
2. Scale cluster to 252 nodes
3. Apply cluster-density e2e-benchmarking workload with 4000 iterations
Actual results:
kube-apiserver CPU usage maxes out at 14 (of 16) cores, and 53.7 GiB (of 124GiB)
Expected results:
kube-apiserver resource consumption scales closer to OpenShiftSDN at this scale, which is 4.52 out of 16 cores and 29.6GiB of 124GiB.
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 (OpenShift Container Platform 4.11.4 bug fix 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/RHBA-2022:6376