Skip to content
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.

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

[Feature/agent_framework] Add Undeploy Model Step #236

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
Original file line number Diff line number Diff line change
Expand Up @@ -73,6 +73,8 @@ private CommonValue() {}
/** The provision workflow thread pool name */
public static final String PROVISION_THREAD_POOL = "opensearch_workflow_provision";

/** Success name field */
public static final String SUCCESS = "success";
/** Index name field */
public static final String INDEX_NAME = "index_name";
/** Type field */
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,114 @@
/*
* Copyright OpenSearch Contributors
* SPDX-License-Identifier: Apache-2.0
*
* The OpenSearch Contributors require contributions made to
* this file be licensed under the Apache-2.0 license or a
* compatible open source license.
*/
package org.opensearch.flowframework.workflow;

import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.opensearch.ExceptionsHelper;
import org.opensearch.OpenSearchException;
import org.opensearch.action.FailedNodeException;
import org.opensearch.core.action.ActionListener;
import org.opensearch.flowframework.exception.FlowFrameworkException;
import org.opensearch.flowframework.util.ParseUtils;
import org.opensearch.ml.client.MachineLearningNodeClient;
import org.opensearch.ml.common.transport.undeploy.MLUndeployModelsResponse;

import java.io.IOException;
import java.util.Collections;
import java.util.List;
import java.util.Map;
import java.util.Set;
import java.util.concurrent.CompletableFuture;
import java.util.stream.Collectors;

import static org.opensearch.flowframework.common.CommonValue.MODEL_ID;
import static org.opensearch.flowframework.common.CommonValue.SUCCESS;

/**
* Step to undeploy model
*/
public class UndeployModelStep implements WorkflowStep {

private static final Logger logger = LogManager.getLogger(UndeployModelStep.class);

private MachineLearningNodeClient mlClient;

static final String NAME = "undeploy_model";

/**
* Instantiate this class
* @param mlClient Machine Learning client to perform the undeploy
*/
public UndeployModelStep(MachineLearningNodeClient mlClient) {
this.mlClient = mlClient;
}

@Override
public CompletableFuture<WorkflowData> execute(
String currentNodeId,
WorkflowData currentNodeInputs,
Map<String, WorkflowData> outputs,
Map<String, String> previousNodeInputs
) throws IOException {
CompletableFuture<WorkflowData> undeployModelFuture = new CompletableFuture<>();

ActionListener<MLUndeployModelsResponse> actionListener = new ActionListener<>() {

@Override
public void onResponse(MLUndeployModelsResponse mlUndeployModelsResponse) {
List<FailedNodeException> failures = mlUndeployModelsResponse.getResponse().failures();
if (failures.isEmpty()) {
undeployModelFuture.complete(
new WorkflowData(
Map.ofEntries(Map.entry(SUCCESS, !mlUndeployModelsResponse.getResponse().hasFailures())),
currentNodeInputs.getWorkflowId(),
currentNodeInputs.getNodeId()
)
);
} else {
List<String> failedNodes = failures.stream().map(FailedNodeException::nodeId).collect(Collectors.toList());
String message = "Failed to undeploy model on nodes " + failedNodes;
logger.error(message);
undeployModelFuture.completeExceptionally(new OpenSearchException(message));
}
}

@Override
public void onFailure(Exception e) {
logger.error("Failed to unldeploy model");
undeployModelFuture.completeExceptionally(new FlowFrameworkException(e.getMessage(), ExceptionsHelper.status(e)));
}
};

Set<String> requiredKeys = Set.of(MODEL_ID);
Set<String> optionalKeys = Collections.emptySet();

try {
Map<String, Object> inputs = ParseUtils.getInputsFromPreviousSteps(
requiredKeys,
optionalKeys,
currentNodeInputs,
outputs,
previousNodeInputs
);

String modelId = inputs.get(MODEL_ID).toString();

mlClient.undeploy(new String[] { modelId }, null, actionListener);
} catch (FlowFrameworkException e) {
undeployModelFuture.completeExceptionally(e);
}
return undeployModelFuture;
}

@Override
public String getName() {
return NAME;
}
}
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,7 @@ public WorkflowStepFactory(
);
stepMap.put(RegisterRemoteModelStep.NAME, new RegisterRemoteModelStep(mlClient, flowFrameworkIndicesHandler));
stepMap.put(DeployModelStep.NAME, new DeployModelStep(mlClient));
stepMap.put(UndeployModelStep.NAME, new UndeployModelStep(mlClient));
stepMap.put(CreateConnectorStep.NAME, new CreateConnectorStep(mlClient, flowFrameworkIndicesHandler));
stepMap.put(DeleteConnectorStep.NAME, new DeleteConnectorStep(mlClient));
stepMap.put(ModelGroupStep.NAME, new ModelGroupStep(mlClient, flowFrameworkIndicesHandler));
Expand Down
8 changes: 8 additions & 0 deletions src/main/resources/mappings/workflow-steps.json
Original file line number Diff line number Diff line change
Expand Up @@ -83,6 +83,14 @@
"deploy_model_status"
]
},
"undeploy_model": {
"inputs":[
"model_id"
],
"outputs":[
"success"
]
},
"register_model_group": {
"inputs":[
"name"
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,131 @@
/*
* Copyright OpenSearch Contributors
* SPDX-License-Identifier: Apache-2.0
*
* The OpenSearch Contributors require contributions made to
* this file be licensed under the Apache-2.0 license or a
* compatible open source license.
*/
package org.opensearch.flowframework.workflow;

import org.opensearch.OpenSearchException;
import org.opensearch.action.FailedNodeException;
import org.opensearch.cluster.ClusterName;
import org.opensearch.core.action.ActionListener;
import org.opensearch.core.rest.RestStatus;
import org.opensearch.flowframework.exception.FlowFrameworkException;
import org.opensearch.ml.client.MachineLearningNodeClient;
import org.opensearch.ml.common.transport.undeploy.MLUndeployModelNodesResponse;
import org.opensearch.ml.common.transport.undeploy.MLUndeployModelsResponse;
import org.opensearch.test.OpenSearchTestCase;

import java.io.IOException;
import java.util.Collections;
import java.util.List;
import java.util.Map;
import java.util.concurrent.CompletableFuture;
import java.util.concurrent.ExecutionException;

import org.mockito.Mock;
import org.mockito.MockitoAnnotations;

import static org.opensearch.flowframework.common.CommonValue.MODEL_ID;
import static org.opensearch.flowframework.common.CommonValue.SUCCESS;
import static org.mockito.ArgumentMatchers.any;
import static org.mockito.Mockito.doAnswer;
import static org.mockito.Mockito.verify;

public class UndeployModelStepTests extends OpenSearchTestCase {
private WorkflowData inputData;

@Mock
MachineLearningNodeClient machineLearningNodeClient;

@Override
public void setUp() throws Exception {
super.setUp();

MockitoAnnotations.openMocks(this);

inputData = new WorkflowData(Collections.emptyMap(), "test-id", "test-node-id");
}

public void testUndeployModel() throws IOException, ExecutionException, InterruptedException {

String modelId = randomAlphaOfLength(5);
UndeployModelStep UndeployModelStep = new UndeployModelStep(machineLearningNodeClient);

doAnswer(invocation -> {
ClusterName clusterName = new ClusterName("clusterName");
ActionListener<MLUndeployModelsResponse> actionListener = invocation.getArgument(2);
MLUndeployModelNodesResponse mlUndeployModelNodesResponse = new MLUndeployModelNodesResponse(
clusterName,
Collections.emptyList(),
Collections.emptyList()
);
MLUndeployModelsResponse output = new MLUndeployModelsResponse(mlUndeployModelNodesResponse);
actionListener.onResponse(output);
return null;
}).when(machineLearningNodeClient).undeploy(any(String[].class), any(), any());

CompletableFuture<WorkflowData> future = UndeployModelStep.execute(
inputData.getNodeId(),
inputData,
Map.of("step_1", new WorkflowData(Map.of(MODEL_ID, modelId), "workflowId", "nodeId")),
Map.of("step_1", MODEL_ID)
);
verify(machineLearningNodeClient).undeploy(any(String[].class), any(), any());

assertTrue(future.isDone());
assertTrue((boolean) future.get().getContent().get(SUCCESS));
}

public void testNoModelIdInOutput() throws IOException {
UndeployModelStep UndeployModelStep = new UndeployModelStep(machineLearningNodeClient);

CompletableFuture<WorkflowData> future = UndeployModelStep.execute(
inputData.getNodeId(),
inputData,
Collections.emptyMap(),
Collections.emptyMap()
);

assertTrue(future.isCompletedExceptionally());
ExecutionException ex = assertThrows(ExecutionException.class, () -> future.get().getContent());
assertTrue(ex.getCause() instanceof FlowFrameworkException);
assertEquals("Missing required inputs [model_id] in workflow [test-id] node [test-node-id]", ex.getCause().getMessage());
}

public void testUndeployModelFailure() throws IOException {
UndeployModelStep UndeployModelStep = new UndeployModelStep(machineLearningNodeClient);

doAnswer(invocation -> {
ClusterName clusterName = new ClusterName("clusterName");
ActionListener<MLUndeployModelsResponse> actionListener = invocation.getArgument(2);
MLUndeployModelNodesResponse mlUndeployModelNodesResponse = new MLUndeployModelNodesResponse(
clusterName,
Collections.emptyList(),
List.of(new FailedNodeException("failed-node", "Test message", null))
);
MLUndeployModelsResponse output = new MLUndeployModelsResponse(mlUndeployModelNodesResponse);
actionListener.onResponse(output);

actionListener.onFailure(new FlowFrameworkException("Failed to undeploy model", RestStatus.INTERNAL_SERVER_ERROR));
return null;
}).when(machineLearningNodeClient).undeploy(any(String[].class), any(), any());

CompletableFuture<WorkflowData> future = UndeployModelStep.execute(
inputData.getNodeId(),
inputData,
Map.of("step_1", new WorkflowData(Map.of(MODEL_ID, "test"), "workflowId", "nodeId")),
Map.of("step_1", MODEL_ID)
);

verify(machineLearningNodeClient).undeploy(any(String[].class), any(), any());

assertTrue(future.isCompletedExceptionally());
ExecutionException ex = assertThrows(ExecutionException.class, () -> future.get().getContent());
assertTrue(ex.getCause() instanceof OpenSearchException);
assertEquals("Failed to undeploy model on nodes [failed-node]", ex.getCause().getMessage());
}
}
Loading