Skip to content

feat: support inferencepool status#979

Merged
mathetake merged 2 commits intoenvoyproxy:mainfrom
Xunzhuo:feat-inference-status
Aug 5, 2025
Merged

feat: support inferencepool status#979
mathetake merged 2 commits intoenvoyproxy:mainfrom
Xunzhuo:feat-inference-status

Conversation

@Xunzhuo
Copy link
Copy Markdown
Member

@Xunzhuo Xunzhuo commented Jul 31, 2025

Description

This PR added support for GIE conformance test and inferencepool status computing.

Now envoy ai gateway have passed all the GIE upstream conformance tests. And it is running in our CI, make sure future changes wont break the conformance.

@Xunzhuo Xunzhuo force-pushed the feat-inference-status branch 9 times, most recently from c24990d to cb0496a Compare August 1, 2025 06:37
@Xunzhuo Xunzhuo force-pushed the feat-inference-status branch from c785de9 to 6dc493f Compare August 1, 2025 08:57
@Xunzhuo Xunzhuo marked this pull request as ready for review August 1, 2025 08:58
@Xunzhuo Xunzhuo requested a review from a team as a code owner August 1, 2025 08:58
@Xunzhuo Xunzhuo added this to the v0.3.0 milestone Aug 1, 2025
Copy link
Copy Markdown
Member

@mathetake mathetake left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is exciting. Thank you for taking a shot

Comment on lines +138 to +148
inferencePoolC := NewInferencePoolController(c, kubernetes.NewForConfigOrDie(config), logger.
WithName("inference-pool"))
if err = TypedControllerBuilderForCRD(mgr, &gwaiev1a2.InferencePool{}).
Watches(&gwapiv1.Gateway{}, inferencePoolC.gatewayEventHandler()).
Watches(&aigv1a1.AIGatewayRoute{}, inferencePoolC.routeEventHandler()).
Watches(&gwapiv1.HTTPRoute{}, inferencePoolC.httpRouteEventHandler()).
Complete(inferencePoolC); err != nil {
return fmt.Errorf("failed to create controller for InferencePool: %w", err)
}
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

does this not fail for the cluster where inferencepool CRD is not installed? If this fails, then shall we make this optional by simply emitting the error log like ("failed to start the inference pool cotroller. This is most likely due to Gateway API Inference Extension CRDs are not installed on the cluster. Please referece to the Envoy AI Gateawy documentation for details."). Alternativel, we can add a flag to the cnotroller like --inferenceExtensionEnabled etc

Signed-off-by: bitliu <bitliu@tencent.com>
@Xunzhuo Xunzhuo force-pushed the feat-inference-status branch 5 times, most recently from 031e779 to 57a046f Compare August 5, 2025 08:01
Signed-off-by: bitliu <bitliu@tencent.com>
@Xunzhuo Xunzhuo force-pushed the feat-inference-status branch from 57a046f to a1fcd08 Compare August 5, 2025 11:27
@mathetake mathetake merged commit 8e2e6b6 into envoyproxy:main Aug 5, 2025
23 checks passed
missBerg pushed a commit to missBerg/ai-gateway that referenced this pull request Dec 20, 2025
**Description**

This PR added support for GIE conformance test and inferencepool status
computing.

Now envoy ai gateway have passed all the GIE upstream conformance tests.
And it is running in our CI, make sure future changes wont break the
conformance.

---------

Signed-off-by: bitliu <bitliu@tencent.com>
Signed-off-by: Erica Hughberg <erica.sundberg.90@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants