Chen, Dylan

> [kubernetes/kubernetes#71233 (comment)](https://github.com/kubernetes/kubernetes/issues/71233#issuecomment-447472125) It seems that huge pages don't work with k8s NOT AGREE WITH YOU, ACTUALLY k8s works fine with mysql with hugepage on

> Perhaps also [#451 (comment)](https://github.com/docker-library/postgres/issues/451#issuecomment-447472044) and other comments there. The background is that i have an APPLE-TO-APPLE workload comparison with hugepage on/off, so i cannot set it to off.

The same issue reproduced on `Ethernet Controller E810-C for QSFP`, all failed from v2.97 to v3.00 ``` root@benchmark-n28hx:/opt/trex-v3.00# $TREX_HOME/t-rex-64 --prom -c 15 -i Starting Scapy server.... Scapy server is started...

New findings for root cause addressing on coredns timeout 

Hey guys, Any workaround suggestions or fixes for this issue?

Currently I can workaround this issue with static ip configured as below, However if dhcp enabled, it always fails. Why? My internal network is under control of a physic switch...

Another new finding is that once I deploy it successfully by above workaround, then I deploy a pod with service exposed by external IP, as long as external IP specified...

Here you can see that the `coredns` and `calico-kube-controllers` pods are always not ready even they are running. ``` kubectl get nodes -A -o wide NAME STATUS ROLES AGE VERSION...

New finding from the route table, it seems VPP created a route on incorrect interface. ``` ip route default via 172.17.28.1 dev ens259f0 proto static 10.96.0.0/12 via 192.168.126.254 dev ens817f1...

Continuous update new findings 1. Disable another port of the interface, only reserve one port, it works. 2. As long as I specify the two ports(ens817f0, ens817f1) with diff subnets(192.168.126.4/24,...