Undefined Description of problem:
Machines created by machine-api in OKD-SCOS 4.14 fail to join the cluster. It appears they pivot successfully and then get stuck at the grub menu.
Version-Release number of selected component (if applicable):
4.14.0-0.okd-scos-2023-10-14-200238
How reproducible:
Seems to be 100%
Steps to Reproduce:
1. Run periodic-ci-openshift-release-master-okd-scos-4.14-e2e-vsphere-ovn which is currently just a rehearsal on https://github.com/openshift/release/pull/44127 2. Observe workers don't get nodes, machines are running fine, nodelink controller complains about not being able to find internal IP for the machines. Pivot is successful from watching console, however the host gets stuck at grub menu, if you press enter the host gives an error about the kernel. 3.
Actual results:
screenshot, grub menu https://drive.google.com/file/d/1C8EOEddfp0cNvyQRu1LLyqUgddytPcPi/view?usp=sharing
screenshot after pressing enter https://drive.google.com/file/d/1_RcARxENriQQCfajw2yt0iuOit0rt8Y1/view?usp=sharing
Expected results:
Nodes join the cluster
Additional info:
This bug was originally filed because node-valid-hostname.service failed on startup, however that doesn't appear to be a fatal error. The pivot completes and the host reboots, however upon reboot the node is stuck at the grub menu. Description updated to reflect this new information. Screenshot on that at https://drive.google.com/file/d/10YUjR81c4VoS0yQEdqZiiJ_D5aIpBYjw/view?usp=sharing Richard Vanderpool coordinated console access and indicates that BIOS mode boots but EFI mode does not.
- is triggered by
-
RHEL-4391 secureboot of centos 9 stream not working
-
- In Progress
-