Bug ID 1682441: After simultaneous VELOS controller RMA, Openshift cluster needs to be reinstalled manually

Last Modified: Jun 19, 2025

Affected Product(s):
F5OS F5OS-C, Velos(all modules)

Known Affected Versions:
F5OS-C 1.6.0, F5OS-C 1.6.1, F5OS-C 1.6.2, F5OS-C 1.8.0, F5OS-C 1.8.1

Opened: Sep 30, 2024

Severity: 3-Major

Symptoms

If both VELOS system controllers are swapped simultaneously, Openshift may not start properly, and the system will not recover or be able to add new blades into the cluster. The openshift log file (/var/log/openshift.log) will show this log message repeating: Restarting openshift origin-node, controllers and api Messages similar to these in /var/log/messages: nodename=controller-1 2024-09-19 17:39:13.686057 C | etcdmain: listen tcp 100.65.3.52:2380: bind: cannot assign requested address nodename=controller-2 2024-09-19 21:00:05.873025 C | etcdmain: listen tcp 100.65.3.51:2380: bind: cannot assign requested address 2024-09-24 06:59:16.591720 I | etcdmain: rejected connection from "100.65.3.52:39400" (error "remote error: tls: bad certificate", ServerName "")

Impact

System remains unhealthy.

Conditions

Both system controllers are replaced simultaneously in a VELOS chassis.

Workaround

This issue should not occur if each controller is replaced one-at-a-time. If both system controllers are swapped simultaneously, then once they have booted up, reinstall Openshift by doing the following: 1. Log into the active VELOS system controller as root, and run: touch /var/omd/CLUSTER_REINSTALL The VELOS system begins the OpenShift cluster reinstallation process. This operation can take 90 minutes or more to complete. 2. In order to check the progress of the rebuild, you can run the following command: tail -F /var/log/openshift.log

Fix Information

None

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips