[Bugfix] Respect explicit deploymentMode when minReplicas=0

[jskswamy]

What this PR does

Fixes the bug where minReplicas: 0 on a component (Engine or Router) ignores the explicit deploymentMode: RawDeployment annotation.

Changes

File	Change
deployment.go	Added shouldInferServerless() helper to centralize logic; updated DetermineEngineDeploymentMode() and DetermineDeploymentModes() signatures
controller.go	Pass global deploymentMode to DetermineDeploymentModes()
ingress/reconciler.go	Get global mode from annotations and pass it
components/engine.go	Updated call to new function signature
utils_test.go	Updated existing test for new signature
deployment_test.go	Added 12 new tests for the fix

Key Implementation

Added a centralized helper to avoid duplicate logic:

// shouldInferServerless checks if minReplicas=0 should trigger Serverless mode.
// Returns false if globalDeploymentMode explicitly blocks it (e.g., RawDeployment).
func shouldInferServerless(minReplicas *int, globalDeploymentMode constants.DeploymentModeType) bool {
    if globalDeploymentMode == constants.RawDeployment {
        return false
    }
    return minReplicas != nil && *minReplicas == 0
}

Why we need it

When users explicitly set deploymentMode: RawDeployment via annotation, they expect that mode to be respected. However, setting minReplicas: 0 on Engine or Router would override this to Serverless mode.

This caused:

1. Unexpected Knative Service creation instead of Deployment
2. Confusing errors when Knative is not installed: no kind is registered for the type v1.Service in scheme
3. User's explicit configuration being silently ignored
4. Inability to use external autoscalers (like KEDA) with scale-to-zero

Before

annotations:
  ome.io/deployment-mode: RawDeployment # Ignored!
spec:
  predictor:
    engine:
      minReplicas: 0 # Triggers Serverless anyway

After

Global deploymentMode: RawDeployment is respected. minReplicas: 0 only triggers Serverless when the global mode is not explicitly set to RawDeployment.

Fixes #445

Checklist

• Tests added/updated (if applicable)
• Docs updated (if applicable) - N/A, internal bug fix
• make fmt passes locally
• make vet passes locally
• make tidy passes locally

[gemini-code-assist]

Summary of Changes

Hello @jskswamy, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses a critical issue where user-defined RawDeployment mode was being silently ignored if a component specified minReplicas: 0, leading to unexpected Serverless deployments. The changes ensure that explicit deployment mode annotations are always honored, preventing misconfigurations, confusing errors, and enabling proper use of external autoscalers with scale-to-zero capabilities.

Highlights

• Bug Fix: Deployment Mode Override: Resolved a bug where minReplicas: 0 on an Engine or Router component would incorrectly override an explicitly set deploymentMode: RawDeployment annotation, forcing Serverless mode.
• Centralized Logic for Serverless Inference: Introduced a new helper function, shouldInferServerless(), to centralize the logic for determining if minReplicas=0 should trigger Serverless mode, ensuring it respects the global deployment mode.
• Updated Function Signatures: Modified the signatures of DetermineEngineDeploymentMode() and DetermineDeploymentModes() to accept a globalDeploymentMode parameter, allowing the system to properly consider explicit global settings.
• Enhanced Test Coverage: Added 12 new tests in deployment_test.go to thoroughly validate the fix and ensure that deployment mode inference correctly respects explicit global settings.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request effectively addresses the bug where minReplicas: 0 incorrectly overrides an explicit deploymentMode: RawDeployment annotation. The introduction of the shouldInferServerless helper function is a great improvement, centralizing the decision logic and making the code cleaner and more maintainable. The function signatures have been updated appropriately to propagate the global deployment mode, and the changes are well-tested with a comprehensive new test suite. I've found one issue in the new test file that will cause a compilation failure, which I've commented on. Otherwise, this is a solid fix.