[BUG] Inconsistent cloud-init behavior with multiple node group configurations

[abhishek-sa1]

Describe the bug
Intermittent Issue Observed:
The cloud-init request occasionally fails to originate from the target node when multiple groups and different nodes are involved in the deployment.

To Reproduce
Steps to reproduce the behavior:

1. Deploy cloud-init in RHEL10 with deployment-recipes playbooks.

Expected behavior
The cloud-init process should initiate a request, and an appropriate response must be provided by the server.

Server (please complete the following information):

• OS: RHEL 10

[synackd]

The cloud-init request occasionally fails to originate from the target node

This is indicative of a deployment issue (client-side) rather than a cloud-init server-side issue. What have you done to eliminate the possibility of the issue being client-side?

[abhishek-sa1]

The issue is not originating from the client side. I’ve verified all the cloud-init configurations, and they appear to be correctly set up. However, we’ve observed that in some instances, cloud-init fails to load after node booted. This is still under investigation.

[alexlovelltroy]

Can you share logs that show the client requesting configurations from the server and the server failing to respond? This behavior is what we would expect if the node fails to properly request cloud-init data from the server.