What happened?

There's a typo in the Cilium values.yaml.j2 template where repository is misspelled as repositry in both certgen and envoy image configurations. This prevents custom image repositories (cilium_hubble_certgen_image_repo, cilium_hubble_envoy_image_repo) from being properly applied, causing the system to fall back to the default quay.io repository.

What did you expect to happen?

https://github.com/kubernetes-sigs/kubespray/blob/v2.28.0/roles/network_plugin/cilium/templates/values.yaml.j2

certgen.image.repositry
envoy:image.repositry

repositry -> repository

How can we reproduce it (as minimally and precisely as possible)?

1. Set up an offline environment with a private registry
   2,. Configure the inventory to use custom image repositories
2. Run Kubespray deployment
3. Observe that Cilium still tries to pull images from quay.io instead of the configured registry_host
4. Check the generated values.yaml and find that the custom repository is not applied due to the typo repositry: instead of repository:

The issue prevents offline deployments from using the configured private registry, causing image pull failures.

OS

Ubuntu 22

Version of Ansible

ansible [core 2.16.14]

Version of Python

Python 3.10.18

Version of Kubespray (commit)

2.28.0

Network plugin used

cilium

Full inventory with variables

ansible -i inventory/sample/inventory.ini

Command used to invoke ansible

ansible-playbook -i inventory/sample/inventory.ini cluster.yml

Output of ansible run

The ansible playbook completed successfully without errors, but Cilium pods in the cluster are not functioning properly due to image pull failures. The pods are trying to pull images from quay.io instead of the configured private registry, causing them to fail to start or run correctly.

Anything else we need to know

No response