CI FAIL: almalinux8-kube-ovn fail because of network

[yankay]

Which jobs are failing:

https://gitlab.com/kargo-ci/kubernetes-sigs-kubespray/-/jobs/4298490280

Which test(s) are failing:

Since when has it been failing:

Testgrid link:

Reason for failure:

Curl between pods is working Fail

Anything else we need to know:

[yankay]

HI @oilbeater @liupeng0518

Would you please help to resolve the issue together :-)
The issue makes the test fail of the kubespray release, #10111

[oilbeater]

@yankay I'm not familiar with the almalinux8 os, is there any cluster that I can ssh to do further debugging?

[yankay]

@yankay I'm not familiar with the almalinux8 os, is there any cluster that I can ssh to do further debugging?

HI @oilbeater , thanks for the replay :-)

The issue has occurred in the CI environment. It has been cleanuped, so we can only just download the log file and the cluster-dump file. I can not ssh into it too.

There are some logs in the download files.

And then , the CI is runed in rocky8 and redhat8, There is the same issue. So the Issue is in the almalinux8/rocky8/redhat8 based on the vargrant. These OS can help to reproduct the issue.

[yankay]

It seems the same as the kubeovn/kube-ovn#2535

[yankay]

HI @oilbeater

It's has been fix by Upgrade kube-ovn to v1.11.5 , https://gitlab.com/kargo-ci/kubernetes-sigs-kubespray/-/jobs/4340555477

Thanks for help :-)

[oilbeater]

HI @oilbeater

It's has been fix by Upgrade kube-ovn to v1.11.5 , https://gitlab.com/kargo-ci/kubernetes-sigs-kubespray/-/jobs/4340555477

Thanks for help :-)

👏 👏 👏