Deployment with short name mode enforcing fails

[rabin-io]

Describe the bug
Trying to deploy Trident 25.10.0 on nighty build of OCP 4.21, will fail with

spec.containers{trident-main}: Failed to inspect image "": rpc error: code = Unknown desc = short name mode is enforcing, but image name netapp/trident:25.10.0 returns ambiguous list

Environment
Provide accurate information about the environment to help us reproduce the issue.

• Trident version: 25.10.0
• Trident installation flags used: tridentctl install
• Container runtime: [e.g. Docker 19.03.1-CE]
• Kubernetes version: v1.34.2
• Kubernetes orchestrator: OpenShift 4.21.0-rc.1
• Kubernetes enabled feature gates:
• OS:
• NetApp backend types:
• Other:

To Reproduce
OCP 4.21 with short name mode enforcing

Expected behavior
Use the full image path to allow the deployment to work.

Additional context

[reederc42]

Hi @rabin-io, we've found the cause and it will be fixed in the 26.02.0 release.

[rabin-io]

Thanks for the update, @reederc42. For others encountering this in the meantime, the issue can be manually resolved by patching the image field in the manifests to use the full image name.

./tridentctl install -n "${TRIDENT_NAMESPACE}" --kubeconfig "${KUBECONFIG}" --generate-custom-yaml

yq -i '(.spec.template.spec.containers[] | select(.name == "trident-main")).image |= "docker.io/" + .' setup/trident-daemonset.yaml
yq -i '(.spec.template.spec.containers[] | select(.name == "trident-main")).image |= "docker.io/" + .' setup/trident-deployment.yaml

./tridentctl install -n "${TRIDENT_NAMESPACE}" --kubeconfig "${KUBECONFIG}" --use-custom-yaml

While OpenShift 4.21 is expected to revert this enforcement until 4.22, using the full image path is definitely the best practice to avoid ambiguity in restricted environments.