-
Bug
-
Resolution: Duplicate
-
Major
-
None
-
4.14
-
Important
-
No
-
SDN Sprint 241, SDN Sprint 242
-
2
-
Approved
-
False
-
Description of problem:
4.14.0-0.nightly-2023-08-28-154013 has a higher pod ready latency(3s +) compared to 4.14 ec4 and 4.13.10(~1s) in node-density (lite) test
Version-Release number of selected component (if applicable):
4.14.0-0.nightly-2023-08-28-154013
How reproducible:
every time
Steps to Reproduce:
1. Install a cluster and scale up to 24 worker nodes, install 3 infra nodes and move monitoring, ingress, registry components to infra nodes. I used vsphere-8 on IBM Cloud. Mastre and worker node type: 8 vCPU, 16G Infra node size: 48 vCPU, 196G 2. Run node-density (lite) test with 245 pod per node 3. Compare the pod ready latency to 4.13.z, and 4.14 ec4
Actual results:
4.14.0-0.nightly-2023-08-28-154013 has a higher pod ready latency compared to 4.14 ec4 and 4.13.10
Expected results:
4.14 should have similar pod ready latency compared to previous release
Additional info:
OCP Version | Flexy Id | Scale Ci Job | Grafana URL | Cloud | Arch Type | Network Type | Worker Count | PODS_PER_NODE | Avg Pod Ready (ms) | P99 Pod Ready (ms) |
4.14.0-0.nightly-2023-08-28-154013 | 229395 | 198 | dac00d0d-ab0e-4b35-bd08-55c856edc742 | vsphere | amd64 | OVN | 24 | 245 | 3153 | 11991 |
4.14.0-0.nightly-2023-08-28-154013 test2 | 229720 | 220 | b81ac8e3-77ef-4e00-8e4c-3c23a4d2b2b6 | vsphere | amd64 | OVN | 24 | 245 | 3527 | 12483 |
4.13.10 | 229585 | 212 | c4bd2ad5-4e1a-48b7-ba98-180098c52c58 | vsphere | amd64 | OVN | 24 | 245 | 1045 | 1973 |
4.13.10 test2 | 229585 | 217 | 64d2df2f-26e3-4060-8276-7271234a8620 | vsphere | amd64 | OVN | 24 | 245 | 1047 | 2050 |
4.14.0-ec.4 | 229721 | 219 | c3866a3b-7338-401b-a8cf-97db49d04d62 | vsphere | amd64 | OVN | 24 | 245 | 962 | 2066 |
must-gathers are uploaded to: OCPBUGS-18389 - Google Drive