Skip to content

[8.x] [ML] Directly call Inference API from Proxy (#127342) #127396

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 25, 2025
Merged

[8.x] [ML] Directly call Inference API from Proxy (#127342) #127396

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 20 additions & 3 deletions ...src/main/java/org/elasticsearch/xpack/inference/action/TransportInferenceActionProxy.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,10 +9,15 @@

import org.elasticsearch.ElasticsearchStatusException;
import org.elasticsearch.action.ActionListener;
import org.elasticsearch.action.ActionRequest;
import org.elasticsearch.action.ActionResponse;
import org.elasticsearch.action.ActionType;
import org.elasticsearch.action.support.ActionFilters;
import org.elasticsearch.action.support.ContextPreservingActionListener;
import org.elasticsearch.action.support.HandledTransportAction;
import org.elasticsearch.client.internal.Client;
import org.elasticsearch.common.util.concurrent.EsExecutors;
import org.elasticsearch.common.util.concurrent.ThreadContext;
import org.elasticsearch.common.xcontent.XContentHelper;
import org.elasticsearch.inference.TaskType;
import org.elasticsearch.inference.UnparsedModel;
Expand All @@ -30,7 +35,6 @@
import java.io.IOException;

import static org.elasticsearch.xpack.core.ClientHelper.INFERENCE_ORIGIN;
import static org.elasticsearch.xpack.core.ClientHelper.executeAsyncWithOrigin;

public class TransportInferenceActionProxy extends HandledTransportAction<InferenceActionProxy.Request, InferenceAction.Response> {
private final ModelRegistry modelRegistry;
Expand Down Expand Up @@ -103,7 +107,7 @@ private void sendUnifiedCompletionRequest(InferenceActionProxy.Request request,
);
}

executeAsyncWithOrigin(client, INFERENCE_ORIGIN, UnifiedCompletionAction.INSTANCE, unifiedRequest, unifiedErrorFormatListener);
execute(UnifiedCompletionAction.INSTANCE, unifiedRequest, listener);
} catch (Exception e) {
unifiedErrorFormatListener.onFailure(e);
}
Expand All @@ -122,6 +126,19 @@ private void sendInferenceActionRequest(InferenceActionProxy.Request request, Ac
inferenceActionRequestBuilder.setInferenceTimeout(request.getTimeout()).setStream(request.isStreaming());
}

executeAsyncWithOrigin(client, INFERENCE_ORIGIN, InferenceAction.INSTANCE, inferenceActionRequestBuilder.build(), listener);
execute(InferenceAction.INSTANCE, inferenceActionRequestBuilder.build(), listener);
}

private <Request extends ActionRequest, Response extends ActionResponse> void execute(
ActionType<Response> action,
Request request,
ActionListener<Response> listener
) {
var threadContext = client.threadPool().getThreadContext();
// stash the context so we clear the user's security headers, then restore and copy the response headers
var supplier = threadContext.newRestorableContext(true);
try (ThreadContext.StoredContext ignore = threadContext.stashWithOrigin(INFERENCE_ORIGIN)) {
client.execute(action, request, new ContextPreservingActionListener<>(supplier, listener));
}
}
}