NIFI-3415: Add Rollback on Failure.

- Added org.apache.nifi.processor.util.pattern package in nifi-processor-utils containing reusable functions to mix-in 'Rollback on Failure' capability.
- Created a process pattern classes, Put and PutGroup. It will be helpful to standardize Processor implementations.
- Applied Rollback on Failure to PutSQL, PutHiveQL, PutHiveStreaming and PutDatabaseRecord.
- Stop using AbstractProcessor for these processors, as it penalizes FlowFiles being processed when it rollback a process session. If FlowFiles are penalized, it will not be fetched again until penalization expires.
- Yield processor when a failure occurs and RollbackOnFailure is enabled. If we do not penalize nor yield, a failed FlowFile retries too frequently.
- When Rollback on Failure is enabled but processor is not transactional, discontinue when an error occurred after successful processes.
- Fixed existing issues on PutHiveStreaming:
  - Output FlowFile Avro format was corrupted by concatenating multiple Avro files.
  - Output FlowFile records had incorrect values because of reusing GenericRecord instance.

Signed-off-by: Matt Burgess <mattyb149@apache.org>

This closes #1658

nifi-commons/nifi-processor-utilities/pom.xml

@@ -53,5 +53,15 @@
             <artifactId>nifi-ssl-context-service-api</artifactId>
             <scope>provided</scope>
         </dependency>
+        <dependency>
+            <groupId>org.apache.nifi</groupId>
+            <artifactId>nifi-mock</artifactId>
+            <scope>test</scope>
+        </dependency>
+        <dependency>
+            <groupId>junit</groupId>
+            <artifactId>junit</artifactId>
+            <scope>test</scope>
+        </dependency>
     </dependencies>
 </project>

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/DiscontinuedException.java

@@ -0,0 +1,31 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+/**
+ * Represents a looping process was discontinued.
+ * When a method throws this exception, its caller should stop processing further inputs and stop immediately.
+ */
+public class DiscontinuedException extends RuntimeException {
+    public DiscontinuedException(String message) {
+        super(message);
+    }
+
+    public DiscontinuedException(String message, Throwable cause) {
+        super(message, cause);
+    }
+}

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/ErrorTypes.java

@@ -0,0 +1,148 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import static org.apache.nifi.processor.util.pattern.ErrorTypes.Destination.Failure;
+import static org.apache.nifi.processor.util.pattern.ErrorTypes.Destination.ProcessException;
+import static org.apache.nifi.processor.util.pattern.ErrorTypes.Destination.Retry;
+import static org.apache.nifi.processor.util.pattern.ErrorTypes.Destination.Self;
+import static org.apache.nifi.processor.util.pattern.ErrorTypes.Penalty.None;
+import static org.apache.nifi.processor.util.pattern.ErrorTypes.Penalty.Penalize;
+import static org.apache.nifi.processor.util.pattern.ErrorTypes.Penalty.Yield;
+
+/**
+ * Represents general error types and how it should be treated.
+ */
+public enum ErrorTypes {
+
+    /**
+     * Procedure setting has to be fixed, otherwise the same error would occur irrelevant to the input.
+     * In order to NOT call failing process frequently, this should be yielded.
+     */
+    PersistentFailure(ProcessException, Yield),
+
+    /**
+     * It is unknown whether the error is persistent or temporal, related to the input or not.
+     */
+    UnknownFailure(ProcessException, None),
+
+    /**
+     * The input will be sent to the failure route for recovery without penalizing.
+     * Basically, the input should not be sent to the same procedure again unless the issue has been solved.
+     */
+    InvalidInput(Failure, None),
+
+    /**
+     * The procedure is temporarily unavailable, usually due to the external service unavailability.
+     * Retrying maybe successful, but it should be yielded for a while.
+     */
+    TemporalFailure(Retry, Yield),
+
+    /**
+     * The input was not processed successfully due to some temporal error
+     * related to the specifics of the input. Retrying maybe successful,
+     * but it should be penalized for a while.
+     */
+    TemporalInputFailure(Retry, Penalize),
+
+    /**
+     * The input was not ready for being processed. It will be kept in the incoming queue and also be penalized.
+     */
+    Defer(Self, Penalize);
+
+    private final Destination destination;
+    private final Penalty penalty;
+    ErrorTypes(Destination destination, Penalty penalty){
+        this.destination = destination;
+        this.penalty = penalty;
+    }
+
+    public Result result() {
+        return new Result(destination, penalty);
+    }
+
+    /**
+     * Represents the destination of input.
+     */
+    public enum Destination {
+        ProcessException, Failure, Retry, Self
+    }
+
+    /**
+     * Indicating yield or penalize the processing when transfer the input.
+     */
+    public enum Penalty {
+        Yield, Penalize, None
+    }
+
+    public Destination destination(){
+        return this.destination;
+    }
+
+    public Penalty penalty(){
+        return this.penalty;
+    }
+
+    /**
+     * Result represents a result of a procedure.
+     * ErrorTypes enum contains basic error result patterns.
+     */
+    public static class Result {
+        private final Destination destination;
+        private final Penalty penalty;
+
+        public Result(Destination destination, Penalty penalty) {
+            this.destination = destination;
+            this.penalty = penalty;
+        }
+
+        public Destination destination() {
+            return destination;
+        }
+
+        public Penalty penalty() {
+            return penalty;
+        }
+
+        @Override
+        public String toString() {
+            return "Result{" +
+                    "destination=" + destination +
+                    ", penalty=" + penalty +
+                    '}';
+        }
+
+        @Override
+        public boolean equals(Object o) {
+            if (this == o) return true;
+            if (o == null || getClass() != o.getClass()) return false;
+
+            Result result = (Result) o;
+
+            if (destination != result.destination) return false;
+            return penalty == result.penalty;
+        }
+
+        @Override
+        public int hashCode() {
+            int result = destination != null ? destination.hashCode() : 0;
+            result = 31 * result + (penalty != null ? penalty.hashCode() : 0);
+            return result;
+        }
+    }
+
+}

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/ExceptionHandler.java

@@ -0,0 +1,235 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.flowfile.FlowFile;
+import org.apache.nifi.logging.ComponentLog;
+import org.apache.nifi.processor.ProcessContext;
+import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.Relationship;
+import org.apache.nifi.processor.exception.ProcessException;
+import org.apache.nifi.processor.util.pattern.ErrorTypes.Result;
+
+import java.util.Collections;
+import java.util.List;
+import java.util.function.BiFunction;
+import java.util.function.Function;
+
+/**
+ * <p>ExceptionHandler provides a structured Exception handling logic composed by reusable partial functions.
+ *
+ * <p>
+ *     Benefits of using ExceptionHandler:
+ *     <li>Externalized error handling code which provides cleaner program only focusing on the expected path.</li>
+ *     <li>Classify specific Exceptions into {@link ErrorTypes}, consolidated error handling based on error type.</li>
+ *     <li>Context aware error handling, {@link RollbackOnFailure} for instance.</li>
+ * </p>
+ */
+public class ExceptionHandler<C> {
+
+    @FunctionalInterface
+    public interface Procedure<I> {
+        void apply(I input) throws Exception;
+    }
+
+    public interface OnError<C, I> {
+        void apply(C context, I input, Result result, Exception e);
+
+        default OnError<C, I> andThen(OnError<C, I> after) {
+            return (c, i, r, e) -> {
+                apply(c, i, r, e);
+                after.apply(c, i, r, e);
+            };
+        }
+    }
+
+    /**
+     * Simply categorise an Exception.
+     */
+    private Function<Exception, ErrorTypes> mapException;
+
+    /**
+     * Adjust error type based on the context.
+     */
+    private BiFunction<C, ErrorTypes, Result> adjustError;
+
+    /**
+     * Do some action to the input based on the final error type.
+     */
+    private OnError<C, ?> onError;
+
+    /**
+     * Specify a function that maps an Exception to certain ErrorType.
+     */
+    public void mapException(Function<Exception, ErrorTypes> mapException) {
+        this.mapException = mapException;
+    }
+
+    /**
+     * <p>Specify a function that adjust ErrorType based on a function context.
+     * <p>For example, {@link RollbackOnFailure#createAdjustError(ComponentLog)} decides
+     * whether a process session should rollback or transfer input to failure or retry.
+     */
+    public void adjustError(BiFunction<C, ErrorTypes, Result> adjustError) {
+        this.adjustError = adjustError;
+    }
+
+    /**
+     * <p>Specify a default OnError function that will be called if one is not explicitly specified when {@link #execute(Object, Object, Procedure)} is called.
+     */
+    public void onError(OnError<C, ?> onError) {
+        this.onError = onError;
+    }
+
+    /**
+     * <p>Executes specified procedure function with the input.
+     * <p>Default OnError function will be called when an exception is thrown.
+     * @param context function context
+     * @param input input for procedure
+     * @param procedure a function that does something with the input
+     * @return True if the procedure finished without issue. False if procedure threw an Exception but it was handled by {@link OnError}.
+     * @throws ProcessException Thrown if the exception was not handled by {@link OnError}
+     * @throws DiscontinuedException Indicating the exception was handled by {@link OnError} but process should stop immediately
+     * without processing any further input
+     */
+    @SuppressWarnings("unchecked")
+    public <I> boolean execute(C context, I input, Procedure<I> procedure) throws ProcessException, DiscontinuedException {
+        return execute(context, input, procedure, (OnError<C, I>) onError);
+    }
+
+    /**
+     * <p>Executes specified procedure function with the input.
+     * @param context function context
+     * @param input input for procedure
+     * @param procedure a function that does something with the input
+     * @param onError specify {@link OnError} function for this execution
+     * @return True if the procedure finished without issue. False if procedure threw an Exception but it was handled by {@link OnError}.
+     * @throws ProcessException Thrown if the exception was not handled by {@link OnError}
+     * @throws DiscontinuedException Indicating the exception was handled by {@link OnError} but process should stop immediately
+     * without processing any further input
+     */
+    public <I> boolean execute(C context, I input, Procedure<I> procedure, OnError<C, I> onError) throws ProcessException, DiscontinuedException {
+        try {
+            procedure.apply(input);
+            return true;
+        } catch (Exception e) {
+
+            if (mapException == null) {
+                throw new ProcessException("An exception was thrown: " + e, e);
+            }
+
+            final ErrorTypes type = mapException.apply(e);
+
+            final Result result;
+            if (adjustError != null) {
+                result = adjustError.apply(context, type);
+            } else {
+                result = new Result(type.destination(), type.penalty());
+            }
+
+            if (onError == null) {
+                throw new IllegalStateException("OnError is not set.");
+            }
+
+            onError.apply(context, input, result, e);
+        }
+        return false;
+    }
+
+    private static FlowFile penalize(final ProcessContext context, final ProcessSession session,
+                                     final FlowFile flowFile, final ErrorTypes.Penalty penalty) {
+        switch (penalty) {
+            case Penalize:
+                return session.penalize(flowFile);
+            case Yield:
+                context.yield();
+        }
+        return flowFile;
+    }
+
+    /**
+     * Create a {@link OnError} function instance that routes input based on {@link Result} destination and penalty.
+     * @param context process context is used to yield a processor
+     * @param session process session is used to penalize a FlowFile
+     * @param routingResult input FlowFile will be routed to a destination relationship in this {@link RoutingResult}
+     * @param relFailure specify failure relationship of a processor
+     * @param relRetry specify retry relationship of a processor
+     * @return composed function
+     */
+    public static <C> ExceptionHandler.OnError<C, FlowFile> createOnError(
+            final ProcessContext context, final ProcessSession session, final RoutingResult routingResult,
+            final Relationship relFailure, final Relationship relRetry) {
+
+        return (fc, input, result, e) -> {
+            final PartialFunctions.FlowFileGroup flowFileGroup = () -> Collections.singletonList(input);
+            createOnGroupError(context, session, routingResult, relFailure, relRetry).apply(fc, flowFileGroup, result, e);
+        };
+    }
+
+    /**
+     * Same as {@link #createOnError(ProcessContext, ProcessSession, RoutingResult, Relationship, Relationship)} for FlowFileGroup.
+     * @param context process context is used to yield a processor
+     * @param session process session is used to penalize FlowFiles
+     * @param routingResult input FlowFiles will be routed to a destination relationship in this {@link RoutingResult}
+     * @param relFailure specify failure relationship of a processor
+     * @param relRetry specify retry relationship of a processor
+     * @return composed function
+     */
+    public static <C, I extends PartialFunctions.FlowFileGroup> ExceptionHandler.OnError<C, I> createOnGroupError(
+            final ProcessContext context, final ProcessSession session, final RoutingResult routingResult,
+            final Relationship relFailure, final Relationship relRetry) {
+        return (c, g, r, e) -> {
+            final Relationship routeTo;
+            switch (r.destination()) {
+                case Failure:
+                    routeTo = relFailure;
+                    break;
+                case Retry:
+                    routeTo = relRetry;
+                    break;
+                case Self:
+                    routeTo = Relationship.SELF;
+                    break;
+                default:
+                    if (e instanceof ProcessException) {
+                        throw (ProcessException)e;
+                    } else {
+                        Object inputs = null;
+                        if (g != null) {
+                            final List<FlowFile> flowFiles = g.getFlowFiles();
+                            switch (flowFiles.size()) {
+                                case 0:
+                                    inputs = "[]";
+                                    break;
+                                case 1:
+                                    inputs = flowFiles.get(0);
+                                    break;
+                                default:
+                                    inputs = String.format("%d FlowFiles including %s", flowFiles.size(), flowFiles.get(0));
+                                    break;
+                            }
+                        }
+                        throw new ProcessException(String.format("Failed to process %s due to %s", inputs, e), e);
+                    }
+            }
+            for (FlowFile f : g.getFlowFiles()) {
+                final FlowFile maybePenalized = penalize(context, session, f, r.penalty());
+                routingResult.routeTo(maybePenalized, routeTo);
+            }
+        };
+    }
+}

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/PartialFunctions.java

@@ -0,0 +1,122 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.flowfile.FlowFile;
+import org.apache.nifi.logging.ComponentLog;
+import org.apache.nifi.processor.ProcessContext;
+import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.ProcessSessionFactory;
+import org.apache.nifi.processor.exception.ProcessException;
+
+import java.util.List;
+
+/**
+ * This class contains various partial functions those are reusable among process patterns.
+ */
+public class PartialFunctions {
+
+    @FunctionalInterface
+    public interface InitConnection<FC, C> {
+        C apply(ProcessContext context, ProcessSession session, FC functionContext) throws ProcessException;
+    }
+
+    @FunctionalInterface
+    public interface FetchFlowFiles<FC> {
+        List<FlowFile> apply(ProcessContext context, ProcessSession session, FC functionContext, RoutingResult result) throws ProcessException;
+    }
+
+    @FunctionalInterface
+    public interface OnCompleted<FC, C> {
+        void apply(ProcessContext context, ProcessSession session, FC functionContext, C connection) throws ProcessException;
+    }
+
+    @FunctionalInterface
+    public interface OnFailed<FC, C> {
+        void apply(ProcessContext context, ProcessSession session, FC functionContext, C connection, Exception e) throws ProcessException;
+    }
+
+    @FunctionalInterface
+    public interface Cleanup<FC, C> {
+        void apply(ProcessContext context, ProcessSession session, FC functionContext, C connection) throws ProcessException;
+    }
+
+    @FunctionalInterface
+    public interface FlowFileGroup {
+        List<FlowFile> getFlowFiles();
+    }
+
+    @FunctionalInterface
+    public interface AdjustRoute<FC> {
+        void apply(ProcessContext context, ProcessSession session, FC functionContext, RoutingResult result) throws ProcessException;
+    }
+
+    @FunctionalInterface
+    public interface TransferFlowFiles<FC> {
+        void apply(ProcessContext context, ProcessSession session, FC functionContext, RoutingResult result) throws ProcessException;
+
+        default TransferFlowFiles<FC> andThen(TransferFlowFiles<FC> after) {
+            return (context, session, functionContext, result) -> {
+                apply(context, session, functionContext, result);
+                after.apply(context, session, functionContext, result);
+            };
+        }
+    }
+
+    public static <FCT> PartialFunctions.FetchFlowFiles<FCT> fetchSingleFlowFile() {
+        return (context, session, functionContext, result) -> session.get(1);
+    }
+
+    public static <FCT> PartialFunctions.TransferFlowFiles<FCT> transferRoutedFlowFiles() {
+        return (context, session, functionContext, result)
+                -> result.getRoutedFlowFiles().forEach(((relationship, routedFlowFiles)
+                -> session.transfer(routedFlowFiles, relationship)));
+    }
+
+    @FunctionalInterface
+    public interface OnTrigger {
+        void execute(ProcessSession session) throws ProcessException;
+    }
+
+    @FunctionalInterface
+    public interface RollbackSession {
+        void rollback(ProcessSession session, Throwable t);
+    }
+
+    /**
+     * <p>This method is identical to what {@link org.apache.nifi.processor.AbstractProcessor#onTrigger(ProcessContext, ProcessSession)} does.</p>
+     * <p>Create a session from ProcessSessionFactory and execute specified onTrigger function, and commit the session if onTrigger finishes successfully.</p>
+     * <p>When an Exception is thrown during execution of the onTrigger, the session will be rollback. FlowFiles being processed will be penalized.</p>
+     */
+    public static void onTrigger(ProcessContext context, ProcessSessionFactory sessionFactory, ComponentLog logger, OnTrigger onTrigger) throws ProcessException {
+        onTrigger(context, sessionFactory, logger, onTrigger, (session, t) -> session.rollback(true));
+    }
+
+    public static void onTrigger(
+            ProcessContext context, ProcessSessionFactory sessionFactory, ComponentLog logger, OnTrigger onTrigger,
+            RollbackSession rollbackSession) throws ProcessException {
+        final ProcessSession session = sessionFactory.createSession();
+        try {
+            onTrigger.execute(session);
+            session.commit();
+        } catch (final Throwable t) {
+            logger.error("{} failed to process due to {}; rolling back session", new Object[]{onTrigger, t});
+            rollbackSession.rollback(session, t);
+            throw t;
+        }
+    }
+}

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/Put.java

@@ -0,0 +1,228 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.flowfile.FlowFile;
+import org.apache.nifi.logging.ComponentLog;
+import org.apache.nifi.processor.ProcessContext;
+import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.Relationship;
+import org.apache.nifi.processor.exception.ProcessException;
+
+import java.util.List;
+import java.util.Objects;
+import java.util.stream.Collectors;
+
+/**
+ * Abstract Put pattern class with a generic onTrigger method structure, composed with various partial functions.
+ * @param <FC> Class of context instance which is passed to each partial functions.
+ *            Lifetime of an function context should be limited for a single onTrigger method.
+ * @param <C> Class of connection to a data storage that this pattern puts data into.
+ */
+public class Put<FC, C extends AutoCloseable> {
+    protected PartialFunctions.InitConnection<FC, C> initConnection;
+    protected PartialFunctions.FetchFlowFiles<FC> fetchFlowFiles = PartialFunctions.fetchSingleFlowFile();
+    protected PutFlowFile<FC, C> putFlowFile;
+    protected PartialFunctions.TransferFlowFiles<FC> transferFlowFiles = PartialFunctions.transferRoutedFlowFiles();
+    protected PartialFunctions.AdjustRoute<FC> adjustRoute;
+    protected PartialFunctions.OnCompleted<FC, C> onCompleted;
+    protected PartialFunctions.OnFailed<FC, C> onFailed;
+    protected PartialFunctions.Cleanup<FC, C> cleanup;
+    protected ComponentLog logger;
+
+    /**
+     * Put fetched FlowFiles to a data storage.
+     * @param context process context passed from a Processor onTrigger.
+     * @param session process session passed from a Processor onTrigger.
+     * @param functionContext function context passed from a Processor onTrigger.
+     * @param connection connection to data storage, established by {@link PartialFunctions.InitConnection}.
+     * @param flowFiles FlowFiles fetched from {@link PartialFunctions.FetchFlowFiles}.
+     * @param result Route incoming FlowFiles if necessary.
+     */
+    protected void putFlowFiles(ProcessContext context, ProcessSession session,
+                                        FC functionContext, C connection, List<FlowFile> flowFiles, RoutingResult result) throws ProcessException {
+        for (FlowFile flowFile : flowFiles) {
+            putFlowFile.apply(context, session, functionContext, connection, flowFile, result);
+        }
+    }
+
+    protected void validateCompositePattern() {
+        Objects.requireNonNull(initConnection, "InitConnection function is required.");
+        Objects.requireNonNull(putFlowFile, "PutFlowFile function is required.");
+        Objects.requireNonNull(transferFlowFiles, "TransferFlowFiles function is required.");
+    }
+
+    /**
+     * <p>Processor using this pattern is expected to call this method from its onTrigger.
+     * <p>Typical usage would be constructing a process pattern instance at a processor method
+     * which is annotated with {@link org.apache.nifi.annotation.lifecycle.OnScheduled},
+     * and use pattern.onTrigger from processor.onTrigger.
+     * <p>{@link PartialFunctions.InitConnection} is required at least. In addition to any functions required by an implementation class.
+     * @param context process context passed from a Processor onTrigger.
+     * @param session process session passed from a Processor onTrigger.
+     * @param functionContext function context should be instantiated per onTrigger call.
+     * @throws ProcessException Each partial function can throw ProcessException if onTrigger should stop immediately.
+     */
+    public void onTrigger(ProcessContext context, ProcessSession session, FC functionContext) throws ProcessException {
+
+        validateCompositePattern();
+
+        final RoutingResult result = new RoutingResult();
+        final List<FlowFile> flowFiles = fetchFlowFiles.apply(context, session, functionContext, result);
+
+        // Transfer FlowFiles if there is any.
+        result.getRoutedFlowFiles().forEach(((relationship, routedFlowFiles) ->
+                session.transfer(routedFlowFiles, relationship)));
+
+        if (flowFiles == null || flowFiles.isEmpty()) {
+            logger.debug("No incoming FlowFiles.");
+            return;
+        }
+
+        try (C connection = initConnection.apply(context, session, functionContext)) {
+
+            try {
+                // Execute the core function.
+                try {
+                    putFlowFiles(context, session, functionContext, connection, flowFiles, result);
+                } catch (DiscontinuedException e) {
+                    // Whether it was an error or semi normal is depends on the implementation and reason why it wanted to discontinue.
+                    // So, no logging is needed here.
+                }
+
+                // Extension point to alter routes.
+                if (adjustRoute != null) {
+                    adjustRoute.apply(context, session, functionContext, result);
+                }
+
+                // Put fetched, but unprocessed FlowFiles back to self.
+                final List<FlowFile> transferredFlowFiles = result.getRoutedFlowFiles().values().stream()
+                        .flatMap(List::stream).collect(Collectors.toList());
+                final List<FlowFile> unprocessedFlowFiles = flowFiles.stream()
+                        .filter(flowFile -> !transferredFlowFiles.contains(flowFile)).collect(Collectors.toList());
+                result.routeTo(unprocessedFlowFiles, Relationship.SELF);
+
+                // OnCompleted processing.
+                if (onCompleted != null) {
+                    onCompleted.apply(context, session, functionContext, connection);
+                }
+
+                // Transfer FlowFiles.
+                transferFlowFiles.apply(context, session, functionContext, result);
+
+            } catch (Exception e) {
+                if (onFailed != null) {
+                    onFailed.apply(context, session, functionContext, connection, e);
+                }
+                throw e;
+            } finally {
+                if (cleanup != null) {
+                    cleanup.apply(context, session, functionContext, connection);
+                }
+            }
+
+        } catch (ProcessException e) {
+            throw e;
+        } catch (Exception e) {
+            // Throw uncaught exception as RuntimeException so that this processor will be yielded.
+            final String msg = String.format("Failed to execute due to %s", e);
+            logger.error(msg, e);
+            throw new RuntimeException(msg, e);
+        }
+
+    }
+
+    /**
+     * Specify an optional function that fetches incoming FlowFIles.
+     * If not specified, single FlowFile is fetched on each onTrigger.
+     * @param f Function to fetch incoming FlowFiles.
+     */
+    public void fetchFlowFiles(PartialFunctions.FetchFlowFiles<FC> f) {
+        fetchFlowFiles = f;
+    }
+
+    /**
+     * Specify a function that establishes a connection to target data storage.
+     * This function will be called when there is valid incoming FlowFiles.
+     * The created connection instance is automatically closed when onTrigger is finished.
+     * @param f Function to initiate a connection to a data storage.
+     */
+    public void initConnection(PartialFunctions.InitConnection<FC, C> f) {
+        initConnection = f;
+    }
+
+    /**
+     * Specify a function that puts an incoming FlowFile to target data storage.
+     * @param f a function to put a FlowFile to target storage.
+     */
+    public void putFlowFile(PutFlowFile<FC, C> f) {
+        this.putFlowFile = f;
+    }
+
+    /**
+     * Specify an optional function that adjust routed FlowFiles before transfer it.
+     * @param f a function to adjust route.
+     */
+    public void adjustRoute(PartialFunctions.AdjustRoute<FC> f) {
+        this.adjustRoute = f;
+    }
+
+    /**
+     * Specify an optional function responsible for transferring routed FlowFiles.
+     * If not specified routed FlowFiles are simply transferred to its destination by default.
+     * @param f a function to transfer routed FlowFiles.
+     */
+    public void transferFlowFiles(PartialFunctions.TransferFlowFiles<FC> f) {
+        this.transferFlowFiles = f;
+    }
+
+    /**
+     * Specify an optional function which will be called if input FlowFiles were successfully put to a target storage.
+     * @param f Function to be called when a put operation finishes successfully.
+     */
+    public void onCompleted(PartialFunctions.OnCompleted<FC, C> f) {
+        onCompleted = f;
+    }
+
+    /**
+     * Specify an optional function which will be called if input FlowFiles failed being put to a target storage.
+     * @param f Function to be called when a put operation failed.
+     */
+    public void onFailed(PartialFunctions.OnFailed<FC, C> f) {
+        onFailed = f;
+    }
+
+    /**
+     * Specify an optional function which will be called in a finally block.
+     * Typically useful when a special cleanup operation is needed for the connection.
+     * @param f Function to be called when a put operation finished regardless of whether it succeeded or not.
+     */
+    public void cleanup(PartialFunctions.Cleanup<FC, C> f) {
+        cleanup = f;
+    }
+
+    public void setLogger(ComponentLog logger) {
+        this.logger = logger;
+    }
+
+    @FunctionalInterface
+    public interface PutFlowFile<FC, C> {
+        void apply(ProcessContext context, ProcessSession session, FC functionContext, C connection,
+                   FlowFile flowFile, RoutingResult result) throws ProcessException;
+    }
+
+}

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/PutGroup.java

@@ -0,0 +1,97 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.flowfile.FlowFile;
+import org.apache.nifi.processor.ProcessContext;
+import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.exception.ProcessException;
+
+import java.util.List;
+import java.util.Objects;
+
+/**
+ * Extended Put pattern capable of handling FlowFile groups.
+ * @param <FC> Function context class.
+ * @param <C> Connection class.
+ * @param <FFG> FlowFileGroup class.
+ */
+public class PutGroup<FC, C extends AutoCloseable, FFG extends PartialFunctions.FlowFileGroup> extends Put<FC, C> {
+
+
+    public PutGroup() {
+        // Just to make a composition valid.
+        this.putFlowFile = (context, session, functionContext, connection, inputFlowFile, result) -> {
+            throw new UnsupportedOperationException();
+        };
+    }
+
+    @FunctionalInterface
+    public interface PutFlowFiles<FC, C, FFG> {
+        void apply(ProcessContext context, ProcessSession session, FC functionContext, C connection,
+                            FFG inputFlowFileGroup, RoutingResult result) throws ProcessException;
+    }
+
+    @Override
+    protected void validateCompositePattern() {
+        super.validateCompositePattern();
+        Objects.requireNonNull(groupFlowFiles, "GroupFlowFiles function is required.");
+    }
+
+    /**
+     * PutGroup does not support PutFileFile function for single FlowFile.
+     * Throws UnsupportedOperationException if called.
+     */
+    @Override
+    public void putFlowFile(PutFlowFile<FC, C> putFlowFile) {
+        throw new UnsupportedOperationException("PutFlowFile can not be used with PutGroup pattern. Specify PutFlowFiles instead.");
+    }
+
+    @FunctionalInterface
+    public interface GroupFlowFiles<FC, C, FFG> {
+        List<FFG> apply(ProcessContext context, ProcessSession session, FC functionContext, C connection, List<FlowFile> flowFiles, RoutingResult result) throws ProcessException;
+    }
+
+    private GroupFlowFiles<FC, C, FFG> groupFlowFiles;
+    private PutFlowFiles<FC, C, FFG> putFlowFiles;
+
+    /**
+     * Specify a function that groups input FlowFiles into FlowFile groups.
+     */
+    public void groupFetchedFlowFiles(GroupFlowFiles<FC, C, FFG> f) {
+        groupFlowFiles = f;
+    }
+
+    /**
+     * Specify a function that puts an input FlowFile group to a target storage using a given connection.
+     */
+    public void putFlowFiles(PutFlowFiles<FC, C, FFG> f) {
+        putFlowFiles = f;
+    }
+
+
+    @Override
+    protected void putFlowFiles(ProcessContext context, ProcessSession session, FC functionContext,
+                               C connection, List<FlowFile> flowFiles, RoutingResult result) throws ProcessException {
+        final List<FFG> flowFileGroups = groupFlowFiles
+                .apply(context, session, functionContext, connection, flowFiles, result);
+
+        for (FFG group : flowFileGroups) {
+            putFlowFiles.apply(context, session, functionContext, connection, group, result);
+        }
+    }
+}

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/RollbackOnFailure.java

@@ -0,0 +1,226 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.components.PropertyDescriptor;
+import org.apache.nifi.flowfile.FlowFile;
+import org.apache.nifi.logging.ComponentLog;
+import org.apache.nifi.processor.ProcessContext;
+import org.apache.nifi.processor.ProcessSessionFactory;
+import org.apache.nifi.processor.Relationship;
+import org.apache.nifi.processor.exception.ProcessException;
+import org.apache.nifi.processor.util.StandardValidators;
+import org.apache.nifi.processor.util.pattern.PartialFunctions.AdjustRoute;
+
+import java.util.List;
+import java.util.Map;
+import java.util.function.BiFunction;
+
+/**
+ * <p>RollbackOnFailure can be used as a function context for process patterns such as {@link Put} to provide a configurable error handling.
+ *
+ * <p>
+ *     RollbackOnFailure can add following characteristics to a processor:
+ *     <li>When disabled, input FlowFiles caused an error will be routed to 'failure' or 'retry' relationship, based on the type of error.</li>
+ *     <li>When enabled, input FlowFiles are kept in the input queue. A ProcessException is thrown to rollback the process session.</li>
+ *     <li>It assumes anything happened during a processors onTrigger can rollback, if this is marked as transactional.</li>
+ *     <li>If transactional and enabled, even if some FlowFiles are already processed, it rollbacks the session when error occurs.</li>
+ *     <li>If not transactional and enabled, it only rollbacks the session when error occurs only if there was no progress.</li>
+ * </p>
+ *
+ * <p>There are two approaches to apply RollbackOnFailure. One is using {@link ExceptionHandler#adjustError(BiFunction)},
+ * and the other is implementing processor onTrigger using process patterns such as {@link Put#adjustRoute(AdjustRoute)}. </p>
+ *
+ * <p>It's also possible to use both approaches. ExceptionHandler can apply when an Exception is thrown immediately, while AdjustRoute respond later but requires less code.</p>
+ */
+public class RollbackOnFailure {
+
+    private final boolean rollbackOnFailure;
+    private final boolean transactional;
+    private boolean discontinue;
+
+    private int processedCount = 0;
+
+    /**
+     * Constructor.
+     * @param rollbackOnFailure Should be set by user via processor configuration.
+     * @param transactional Specify whether a processor is transactional.
+     *                      If not, it is important to call {@link #proceed()} after successful execution of processors task,
+     *                      that indicates processor made an operation that can not be undone.
+     */
+    public RollbackOnFailure(boolean rollbackOnFailure, boolean transactional) {
+        this.rollbackOnFailure = rollbackOnFailure;
+        this.transactional = transactional;
+    }
+
+    public static final PropertyDescriptor ROLLBACK_ON_FAILURE = createRollbackOnFailureProperty("");
+
+    public static  PropertyDescriptor createRollbackOnFailureProperty(String additionalDescription) {
+        return new PropertyDescriptor.Builder()
+                .name("rollback-on-failure")
+                .displayName("Rollback On Failure")
+                .description("Specify how to handle error." +
+                        " By default (false), if an error occurs while processing a FlowFile, the FlowFile will be routed to" +
+                        " 'failure' or 'retry' relationship based on error type, and processor can continue with next FlowFile." +
+                        " Instead, you may want to rollback currently processed FlowFiles and stop further processing immediately." +
+                        " In that case, you can do so by enabling this 'Rollback On Failure' property. " +
+                        " If enabled, failed FlowFiles will stay in the input relationship without penalizing it and being processed repeatedly" +
+                        " until it gets processed successfully or removed by other means." +
+                        " It is important to set adequate 'Yield Duration' to avoid retrying too frequently." + additionalDescription)
+                .allowableValues("true", "false")
+                .addValidator(StandardValidators.BOOLEAN_VALIDATOR)
+                .defaultValue("false")
+                .required(true)
+                .build();
+    }
+
+    /**
+     * Create a function to use with {@link ExceptionHandler} that adjust error type based on functional context.
+     */
+    public static <FCT extends RollbackOnFailure> BiFunction<FCT, ErrorTypes, ErrorTypes.Result> createAdjustError(final ComponentLog logger) {
+        return (c, t) -> {
+
+            ErrorTypes.Result adjusted = null;
+            switch (t.destination()) {
+
+                case ProcessException:
+                    // If this process can rollback, then rollback it.
+                    if (!c.canRollback()) {
+                        // If an exception is thrown but the processor is not transactional and processed count > 0, adjust it to self,
+                        // in order to stop any further processing until this input is processed successfully.
+                        // If we throw an Exception in this state, the already succeeded FlowFiles will be rolled back, too.
+                        // In case the progress was made by other preceding inputs,
+                        // those successful inputs should be sent to 'success' and this input stays in incoming queue.
+                        // In case this input made some progress to external system, the partial update will be replayed again,
+                        // can cause duplicated data.
+                        c.discontinue();
+                        // We should not penalize a FlowFile, if we did, other FlowFiles can be fetched first.
+                        // We need to block others to be processed until this one finishes.
+                        adjusted = new ErrorTypes.Result(ErrorTypes.Destination.Self, ErrorTypes.Penalty.Yield);
+                    }
+                    break;
+
+                case Failure:
+                case Retry:
+                    if (c.isRollbackOnFailure()) {
+                        c.discontinue();
+                        if (c.canRollback()) {
+                            // If this process can rollback, then throw ProcessException instead, in order to rollback.
+                            adjusted = new ErrorTypes.Result(ErrorTypes.Destination.ProcessException, ErrorTypes.Penalty.Yield);
+                        } else {
+                            // If not,
+                            adjusted = new ErrorTypes.Result(ErrorTypes.Destination.Self, ErrorTypes.Penalty.Yield);
+                        }
+                    }
+                    break;
+            }
+
+            if (adjusted != null) {
+                if (logger.isDebugEnabled()) {
+                    logger.debug("Adjusted {} to {} based on context rollbackOnFailure={}, processedCount={}, transactional={}",
+                            new Object[]{t, adjusted, c.isRollbackOnFailure(), c.getProcessedCount(), c.isTransactional()});
+                }
+                return adjusted;
+            }
+
+            return t.result();
+        };
+    }
+
+    /**
+     * Create an {@link AdjustRoute} function to use with process pattern such as {@link Put} that adjust routed FlowFiles based on context.
+     * This function works as a safety net by covering cases that Processor implementation did not use ExceptionHandler and transfer FlowFiles
+     * without considering RollbackOnFailure context.
+     */
+    public static <FCT extends RollbackOnFailure> AdjustRoute<FCT> createAdjustRoute(Relationship ... failureRelationships) {
+        return (context, session, fc, result) -> {
+            if (fc.isRollbackOnFailure()) {
+                // Check if route contains failure relationship.
+                for (Relationship failureRelationship : failureRelationships) {
+                    if (!result.contains(failureRelationship)) {
+                        continue;
+                    }
+                    if (fc.canRollback()) {
+                        throw new ProcessException(String.format(
+                                "A FlowFile is routed to %s. Rollback session based on context rollbackOnFailure=%s, processedCount=%d, transactional=%s",
+                                failureRelationship.getName(), fc.isRollbackOnFailure(), fc.getProcessedCount(), fc.isTransactional()));
+                    } else {
+                        // Send failed FlowFiles to self.
+                        final Map<Relationship, List<FlowFile>> routedFlowFiles = result.getRoutedFlowFiles();
+                        final List<FlowFile> failedFlowFiles = routedFlowFiles.remove(failureRelationship);
+                        result.routeTo(failedFlowFiles, Relationship.SELF);
+                    }
+                }
+            }
+        };
+    }
+
+    public static <FCT extends RollbackOnFailure, I> ExceptionHandler.OnError<FCT, I> createOnError(ExceptionHandler.OnError<FCT, I> onError) {
+        return onError.andThen((context, input, result, e) -> {
+            if (context.shouldDiscontinue()) {
+                throw new DiscontinuedException("Discontinue processing due to " + e, e);
+            }
+        });
+    }
+
+    public static <FCT extends RollbackOnFailure> void onTrigger(
+            ProcessContext context, ProcessSessionFactory sessionFactory, FCT functionContext, ComponentLog logger,
+            PartialFunctions.OnTrigger onTrigger) throws ProcessException {
+
+        PartialFunctions.onTrigger(context, sessionFactory, logger, onTrigger, (session, t) -> {
+            // If RollbackOnFailure is enabled, do not penalize processing FlowFiles when rollback,
+            // in order to keep those in the incoming relationship to be processed again.
+            final boolean shouldPenalize = !functionContext.isRollbackOnFailure();
+            session.rollback(shouldPenalize);
+
+            // However, keeping failed FlowFile in the incoming relationship would retry it too often.
+            // So, administratively yield the process.
+            if (functionContext.isRollbackOnFailure()) {
+                logger.warn("Administratively yielding {} after rolling back due to {}", new Object[]{context.getName(), t}, t);
+                context.yield();
+            }
+        });
+    }
+
+    public int proceed() {
+        return ++processedCount;
+    }
+
+    public int getProcessedCount() {
+        return processedCount;
+    }
+
+    public boolean isRollbackOnFailure() {
+        return rollbackOnFailure;
+    }
+
+    public boolean isTransactional() {
+        return transactional;
+    }
+
+    public boolean canRollback() {
+        return transactional || processedCount == 0;
+    }
+
+    public boolean shouldDiscontinue() {
+        return discontinue;
+    }
+
+    public void discontinue() {
+        this.discontinue = true;
+    }
+}

nifi-commons/nifi-processor-utilities/src/main/java/org/apache/nifi/processor/util/pattern/RoutingResult.java

@@ -0,0 +1,50 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.flowfile.FlowFile;
+import org.apache.nifi.processor.Relationship;
+
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
+public class RoutingResult {
+
+    private final Map<Relationship, List<FlowFile>> routedFlowFiles = new HashMap<>();
+
+    public void routeTo(final FlowFile flowFile, final Relationship relationship) {
+        routedFlowFiles.computeIfAbsent(relationship, r -> new ArrayList<>()).add(flowFile);
+    }
+
+    public void routeTo(final List<FlowFile> flowFiles, final Relationship relationship) {
+        routedFlowFiles.computeIfAbsent(relationship, r -> new ArrayList<>()).addAll(flowFiles);
+    }
+
+    public void merge(final RoutingResult r) {
+        r.getRoutedFlowFiles().forEach((relationship, routedFlowFiles) -> routeTo(routedFlowFiles, relationship));
+    }
+
+    public Map<Relationship, List<FlowFile>> getRoutedFlowFiles() {
+        return routedFlowFiles;
+    }
+
+    public boolean contains(Relationship relationship) {
+        return routedFlowFiles.containsKey(relationship) && !routedFlowFiles.get(relationship).isEmpty();
+    }
+}

nifi-commons/nifi-processor-utilities/src/test/java/org/apache/nifi/processor/util/pattern/TestExceptionHandler.java

@@ -0,0 +1,202 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.processor.exception.ProcessException;
+import org.junit.Test;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.IOException;
+import java.util.Arrays;
+import java.util.concurrent.atomic.AtomicReference;
+import java.util.function.Function;
+
+import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertTrue;
+import static org.junit.Assert.fail;
+
+public class TestExceptionHandler {
+
+    private static final Logger logger = LoggerFactory.getLogger(TestExceptionHandler.class);
+
+    /**
+     * Simulate an external procedure.
+     */
+    static class ExternalProcedure {
+        private boolean available = true;
+        int divide(Integer a, Integer b) throws Exception {
+            if (!available) {
+                throw new IOException("Not available");
+            }
+            if (a == 10) {
+                throw new IllegalStateException("Service for 10 is not currently available.");
+            }
+            return a / b;
+        }
+    }
+
+    private class Context {
+        int count = 0;
+    }
+
+    @Test
+    public void testBasicUsage() {
+
+        final ExternalProcedure p = new ExternalProcedure();
+
+        try {
+            // Although a catch-exception has to be caught each possible call,
+            // usually the error handling logic will be the same.
+            // Ends up having a lot of same code.
+            final int r1 = p.divide(4, 2);
+            assertEquals(2, r1);
+        } catch (Exception e) {
+            e.printStackTrace();
+        }
+
+        final Context context = new Context();
+        final ExceptionHandler<Context> handler = new ExceptionHandler<>();
+
+        // Using handler can avoid the try catch block with reusable error handling logic.
+        handler.execute(context, 6, i -> {
+            final int r2 = p.divide(i, 2);
+            assertEquals(3, r2);
+        });
+
+        // If return value is needed, use AtomicReference.
+        AtomicReference<Integer> r = new AtomicReference<>();
+        handler.execute(context, 8, i -> r.set(p.divide(i, 2)));
+        assertEquals(4, r.get().intValue());
+
+        // If no exception mapping is specified, any Exception thrown is wrapped by ProcessException.
+        try {
+            final Integer nullInput = null;
+            handler.execute(context, nullInput, i -> r.set(p.divide(i, 2)));
+            fail("Exception should be thrown because input is null.");
+        } catch (ProcessException e) {
+            assertTrue(e.getCause() instanceof NullPointerException);
+        }
+    }
+
+    // Reusable Exception mapping function.
+    static Function<Exception, ErrorTypes> exceptionMapping = i -> {
+        try {
+            throw i;
+        } catch (NullPointerException | ArithmeticException | NumberFormatException e) {
+            return ErrorTypes.InvalidInput;
+        } catch (IllegalStateException e) {
+            return ErrorTypes.TemporalInputFailure;
+        } catch (IOException e) {
+            return ErrorTypes.TemporalFailure;
+        } catch (Exception e) {
+            throw new ProcessException(e);
+        }
+    };
+
+    @Test
+    public void testHandling() {
+
+        final ExternalProcedure p = new ExternalProcedure();
+        final Context context = new Context();
+
+        final ExceptionHandler<Context> handler = new ExceptionHandler<>();
+        handler.mapException(exceptionMapping);
+        handler.onError(createInputErrorHandler());
+
+        // Benefit of handler is being able to externalize error handling, make it simpler.
+        handler.execute(context, 4, i -> {
+            final int r = p.divide(i, 2);
+            assertEquals(2, r);
+        });
+
+        // Null pointer exception.
+        final Integer input = null;
+        handler.execute(context, input, i -> {
+            p.divide(i, 2);
+            fail("Shouldn't reach here.");
+        });
+
+        // Divide by zero.
+        handler.execute(context, 0, i -> {
+            p.divide(2, i);
+            fail("Shouldn't reach here.");
+        });
+
+
+    }
+
+    static <C> ExceptionHandler.OnError<C, Integer> createInputErrorHandler() {
+        return (c, i, r, e) -> {
+            switch (r.destination()) {
+                case ProcessException:
+                    throw new ProcessException(String.format("Execution failed due to %s", e), e);
+                default:
+                    logger.warn(String.format("Routing to %s: %d caused %s", r, i, e));
+            }
+        };
+    }
+
+    static <C> ExceptionHandler.OnError<C, Integer[]> createArrayInputErrorHandler() {
+        return (c, i, r, e) -> {
+            switch (r.destination()) {
+                case ProcessException:
+                    throw new ProcessException(String.format("Execution failed due to %s", e), e);
+                default:
+                    logger.warn(String.format("Routing to %s: %d, %d caused %s", r, i[0], i[1], e));
+            }
+        };
+    }
+
+    @Test
+    public void testHandlingLoop() {
+
+        final ExternalProcedure p = new ExternalProcedure();
+        final Context context = new Context();
+
+        final ExceptionHandler<Context> handler = new ExceptionHandler<>();
+        handler.mapException(exceptionMapping);
+        handler.onError(createArrayInputErrorHandler());
+
+        // It's especially handy when looping through inputs. [a, b, expected result]
+        Integer[][] inputs = new Integer[][]{{4, 2, 2}, {null, 2, 999}, {2, 0, 999}, {10, 2, 999}, {8, 2, 4}};
+
+        Arrays.stream(inputs).forEach(input ->  handler.execute(context, input, (in) -> {
+            final Integer r = p.divide(in[0], in[1]);
+            // This is safe because if p.divide throws error, this code won't be executed.
+            assertEquals(in[2], r);
+        }));
+
+        AtomicReference<Integer> r = new AtomicReference<>();
+        for (Integer[] input : inputs) {
+
+            if (!handler.execute(context, input, (in) -> {
+                r.set(p.divide(in[0], in[1]));
+                context.count++;
+            })){
+                // Handler returns false when it fails.
+                // Cleaner if-exception-continue-next-input can be written cleaner.
+                continue;
+            }
+
+            assertEquals(input[2], r.get());
+        }
+
+        assertEquals("Successful inputs", 2, context.count);
+    }
+
+}

nifi-commons/nifi-processor-utilities/src/test/java/org/apache/nifi/processor/util/pattern/TestRollbackOnFailure.java

@@ -0,0 +1,144 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.processor.util.pattern;
+
+import org.apache.nifi.logging.ComponentLog;
+import org.apache.nifi.processor.exception.ProcessException;
+import org.apache.nifi.processor.util.pattern.TestExceptionHandler.ExternalProcedure;
+import org.apache.nifi.util.MockComponentLog;
+import org.junit.Test;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import static org.apache.nifi.processor.util.pattern.TestExceptionHandler.createArrayInputErrorHandler;
+import static org.apache.nifi.processor.util.pattern.TestExceptionHandler.exceptionMapping;
+import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.fail;
+
+public class TestRollbackOnFailure {
+
+    private static final Logger logger = LoggerFactory.getLogger(TestRollbackOnFailure.class);
+
+    /**
+     * This can be an example for how to compose an ExceptionHandler instance by reusable functions.
+     * @param logger used to log messages within functions
+     * @return a composed ExceptionHandler
+     */
+    private ExceptionHandler<RollbackOnFailure> getContextAwareExceptionHandler(ComponentLog logger) {
+        final ExceptionHandler<RollbackOnFailure> handler = new ExceptionHandler<>();
+        handler.mapException(exceptionMapping);
+        handler.adjustError(RollbackOnFailure.createAdjustError(logger));
+        handler.onError(createArrayInputErrorHandler());
+        return handler;
+    }
+
+    private void processInputs(RollbackOnFailure context, Integer[][] inputs, List<Integer> results) {
+        final ExternalProcedure p = new ExternalProcedure();
+        final MockComponentLog componentLog = new MockComponentLog("processor-id", this);
+        final ExceptionHandler<RollbackOnFailure> handler = getContextAwareExceptionHandler(componentLog);
+
+        for (Integer[] input : inputs) {
+
+            if (!handler.execute(context, input, (in) -> {
+                results.add(p.divide(in[0], in[1]));
+                context.proceed();
+            })){
+                continue;
+            }
+
+            assertEquals(input[2], results.get(results.size() - 1));
+        }
+    }
+
+    @Test
+    public void testContextDefaultBehavior() {
+
+        // Disabling rollbackOnFailure would route Failure or Retry as they are.
+        final RollbackOnFailure context = new RollbackOnFailure(false, false);
+
+        Integer[][] inputs = new Integer[][]{{null, 2, 999}, {4, 2, 2}, {2, 0, 999}, {10, 2, 999}, {8, 2, 4}};
+
+        final List<Integer> results = new ArrayList<>();
+        try {
+            processInputs(context, inputs, results);
+        } catch (ProcessException e) {
+            fail("ProcessException should NOT be thrown");
+        }
+
+        assertEquals("Successful inputs", 2, context.getProcessedCount());
+    }
+
+    @Test
+    public void testContextRollbackOnFailureNonTransactionalFirstFailure() {
+
+        final RollbackOnFailure context = new RollbackOnFailure(true, false);
+
+        // If the first execution fails without any succeeded inputs, it should throw a ProcessException.
+        Integer[][] inputs = new Integer[][]{{null, 2, 999}, {4, 2, 2}, {2, 0, 999}, {10, 2, 999}, {8, 2, 4}};
+
+        final List<Integer> results = new ArrayList<>();
+        try {
+            processInputs(context, inputs, results);
+            fail("ProcessException should be thrown");
+        } catch (ProcessException e) {
+            logger.info("Exception was thrown as expected.");
+        }
+
+        assertEquals("Successful inputs", 0, context.getProcessedCount());
+    }
+
+    @Test
+    public void testContextRollbackOnFailureNonTransactionalAlreadySucceeded() {
+
+        final RollbackOnFailure context = new RollbackOnFailure(true, false);
+
+        // If an execution fails after succeeded inputs, it transfer the input to Failure instead of ProcessException,
+        // and keep going. Because the external system does not support transaction.
+        Integer[][] inputs = new Integer[][]{{4, 2, 2}, {2, 0, 999}, {null, 2, 999}, {10, 2, 999}, {8, 2, 4}};
+
+        final List<Integer> results = new ArrayList<>();
+        try {
+            processInputs(context, inputs, results);
+        } catch (ProcessException e) {
+            fail("ProcessException should NOT be thrown");
+        }
+
+        assertEquals("Successful inputs", 2, context.getProcessedCount());
+    }
+
+    @Test
+    public void testContextRollbackOnFailureTransactionalAlreadySucceeded() {
+
+        final RollbackOnFailure context = new RollbackOnFailure(true, true);
+
+        // Even if an execution fails after succeeded inputs, it transfer the input to Failure,
+        // because the external system supports transaction.
+        Integer[][] inputs = new Integer[][]{{4, 2, 2}, {2, 0, 999}, {null, 2, 999}, {10, 2, 999}, {8, 2, 4}};
+
+        final List<Integer> results = new ArrayList<>();
+        try {
+            processInputs(context, inputs, results);
+            fail("ProcessException should be thrown");
+        } catch (ProcessException e) {
+            logger.info("Exception was thrown as expected.");
+        }
+
+    }
+}

nifi-nar-bundles/nifi-hive-bundle/nifi-hive-processors/pom.xml

@@ -30,6 +30,10 @@
             <groupId>org.apache.nifi</groupId>
             <artifactId>nifi-api</artifactId>
         </dependency>
+        <dependency>
+            <groupId>org.apache.nifi</groupId>
+            <artifactId>nifi-processor-utils</artifactId>
+        </dependency>
         <dependency>
             <groupId>org.apache.nifi</groupId>
             <artifactId>nifi-dbcp-service-api</artifactId>

nifi-nar-bundles/nifi-hive-bundle/nifi-hive-processors/src/main/java/org/apache/nifi/processors/hive/AbstractHiveQLProcessor.java

@@ -19,9 +19,8 @@ package org.apache.nifi.processors.hive;
 import org.apache.nifi.components.PropertyDescriptor;
 import org.apache.nifi.dbcp.hive.HiveDBCPService;
 import org.apache.nifi.flowfile.FlowFile;
-import org.apache.nifi.processor.AbstractProcessor;
+import org.apache.nifi.processor.AbstractSessionFactoryProcessor;
 import org.apache.nifi.processor.ProcessSession;
-import org.apache.nifi.processor.exception.ProcessException;
 import org.apache.nifi.processor.io.InputStreamCallback;
 import org.apache.nifi.processor.util.StandardValidators;
 import org.apache.nifi.stream.io.StreamUtils;
@@ -30,6 +29,7 @@ import java.io.IOException;
 import java.io.InputStream;
 import java.math.BigDecimal;
 import java.nio.charset.Charset;
+import java.sql.SQLDataException;
 import java.sql.Time;
 import java.sql.Timestamp;
 import java.sql.Date;
@@ -45,7 +45,7 @@ import java.util.regex.Pattern;
 /**
  * An abstract base class for HiveQL processors to share common data, methods, etc.
  */
-public abstract class AbstractHiveQLProcessor extends AbstractProcessor {
+public abstract class AbstractHiveQLProcessor extends AbstractSessionFactoryProcessor {
 
     protected static final Pattern HIVEQL_TYPE_ATTRIBUTE_PATTERN = Pattern.compile("hiveql\\.args\\.(\\d+)\\.type");
     protected static final Pattern NUMBER_PATTERN = Pattern.compile("-?\\d+");
@@ -112,7 +112,7 @@ public abstract class AbstractHiveQLProcessor extends AbstractProcessor {
                     if (parameterIndex >= base && parameterIndex < base + paramCount) {
                         final boolean isNumeric = NUMBER_PATTERN.matcher(entry.getValue()).matches();
                         if (!isNumeric) {
-                            throw new ProcessException("Value of the " + key + " attribute is '" + entry.getValue() + "', which is not a valid JDBC numeral jdbcType");
+                            throw new SQLDataException("Value of the " + key + " attribute is '" + entry.getValue() + "', which is not a valid JDBC numeral jdbcType");
                         }
 
                         final String valueAttrName = "hiveql.args." + parameterIndex + ".value";
@@ -139,7 +139,7 @@ public abstract class AbstractHiveQLProcessor extends AbstractProcessor {
             try {
                 setParameter(stmt, ph.attributeName, index, ph.value, ph.jdbcType);
             } catch (final NumberFormatException nfe) {
-                throw new ProcessException("The value of the " + ph.attributeName + " is '" + ph.value + "', which cannot be converted into the necessary data jdbcType", nfe);
+                throw new SQLDataException("The value of the " + ph.attributeName + " is '" + ph.value + "', which cannot be converted into the necessary data jdbcType", nfe);
             }
         }
         return base + paramCount;

nifi-nar-bundles/nifi-hive-bundle/nifi-hive-processors/src/main/java/org/apache/nifi/processors/hive/PutHiveQL.java

@@ -24,14 +24,24 @@ import org.apache.nifi.annotation.behavior.ReadsAttributes;
 import org.apache.nifi.annotation.documentation.CapabilityDescription;
 import org.apache.nifi.annotation.documentation.SeeAlso;
 import org.apache.nifi.annotation.documentation.Tags;
+import org.apache.nifi.annotation.lifecycle.OnScheduled;
 import org.apache.nifi.components.PropertyDescriptor;
 import org.apache.nifi.dbcp.hive.HiveDBCPService;
 import org.apache.nifi.flowfile.FlowFile;
 import org.apache.nifi.processor.ProcessContext;
 import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.ProcessSessionFactory;
 import org.apache.nifi.processor.Relationship;
 import org.apache.nifi.processor.exception.ProcessException;
 import org.apache.nifi.processor.util.StandardValidators;
+import org.apache.nifi.processor.util.pattern.ErrorTypes;
+import org.apache.nifi.processor.util.pattern.ExceptionHandler;
+import org.apache.nifi.processor.util.pattern.ExceptionHandler.OnError;
+import org.apache.nifi.processor.util.pattern.PartialFunctions.FetchFlowFiles;
+import org.apache.nifi.processor.util.pattern.PartialFunctions.InitConnection;
+import org.apache.nifi.processor.util.pattern.Put;
+import org.apache.nifi.processor.util.pattern.RollbackOnFailure;
+import org.apache.nifi.processor.util.pattern.RoutingResult;
 
 import java.nio.charset.Charset;
 import java.sql.Connection;
@@ -108,6 +118,7 @@ public class PutHiveQL extends AbstractHiveQLProcessor {
         _propertyDescriptors.add(BATCH_SIZE);
         _propertyDescriptors.add(CHARSET);
         _propertyDescriptors.add(STATEMENT_DELIMITER);
+        _propertyDescriptors.add(RollbackOnFailure.ROLLBACK_ON_FAILURE);
         propertyDescriptors = Collections.unmodifiableList(_propertyDescriptors);
 
         Set<Relationship> _relationships = new HashSet<>();
@@ -117,6 +128,31 @@ public class PutHiveQL extends AbstractHiveQLProcessor {
         relationships = Collections.unmodifiableSet(_relationships);
     }
 
+    private Put<FunctionContext, Connection> process;
+    private ExceptionHandler<FunctionContext> exceptionHandler;
+
+    @OnScheduled
+    public void constructProcess() {
+        exceptionHandler = new ExceptionHandler<>();
+        exceptionHandler.mapException(e -> {
+            if (e instanceof SQLNonTransientException) {
+                return ErrorTypes.InvalidInput;
+            } else if (e instanceof SQLException) {
+                return ErrorTypes.TemporalFailure;
+            } else {
+                return ErrorTypes.UnknownFailure;
+            }
+        });
+        exceptionHandler.adjustError(RollbackOnFailure.createAdjustError(getLogger()));
+
+        process = new Put<>();
+        process.setLogger(getLogger());
+        process.initConnection(initConnection);
+        process.fetchFlowFiles(fetchFlowFiles);
+        process.putFlowFile(putFlowFile);
+        process.adjustRoute(RollbackOnFailure.createAdjustRoute(REL_FAILURE, REL_RETRY));
+    }
+
     @Override
     protected List<PropertyDescriptor> getSupportedPropertyDescriptors() {
         return propertyDescriptors;
@@ -127,75 +163,95 @@ public class PutHiveQL extends AbstractHiveQLProcessor {
         return relationships;
     }
 
-    @Override
+    private class FunctionContext extends RollbackOnFailure {
-    public void onTrigger(final ProcessContext context, final ProcessSession session) throws ProcessException {
+        final Charset charset;
-        final int batchSize = context.getProperty(BATCH_SIZE).asInteger();
+        final String statementDelimiter;
-        final List<FlowFile> flowFiles = session.get(batchSize);
-
-        if (flowFiles.isEmpty()) {
-            return;
-        }
-
         final long startNanos = System.nanoTime();
-        final Charset charset = Charset.forName(context.getProperty(CHARSET).getValue());
-        final HiveDBCPService dbcpService = context.getProperty(HIVE_DBCP_SERVICE).asControllerService(HiveDBCPService.class);
-        final String statementDelimiter =   context.getProperty(STATEMENT_DELIMITER).getValue();
 
-        try (final Connection conn = dbcpService.getConnection()) {
+        String connectionUrl;
 
-            for (FlowFile flowFile : flowFiles) {
-                try {
-                    final String script = getHiveQL(session, flowFile, charset);
-                    String regex = "(?<!\\\\)" + Pattern.quote(statementDelimiter);
 
-                    String[] hiveQLs = script.split(regex);
+        private FunctionContext(boolean rollbackOnFailure, Charset charset, String statementDelimiter) {
-
+            super(rollbackOnFailure, false);
-                    int loc = 1;
+            this.charset = charset;
-                    for (String hiveQL: hiveQLs) {
+            this.statementDelimiter = statementDelimiter;
-                        getLogger().debug("HiveQL: {}", new Object[]{hiveQL});
-
-                        if (!StringUtils.isEmpty(hiveQL.trim())) {
-                            final PreparedStatement stmt = conn.prepareStatement(hiveQL.trim());
-
-                            // Get ParameterMetadata
-                            // Hive JDBC Doesn't support this yet:
-                            // ParameterMetaData pmd = stmt.getParameterMetaData();
-                            // int paramCount = pmd.getParameterCount();
-
-                            int paramCount = StringUtils.countMatches(hiveQL, "?");
-
-                            if (paramCount > 0) {
-                                loc = setParameters(loc, stmt, paramCount, flowFile.getAttributes());
-                            }
-
-                            // Execute the statement
-                            stmt.execute();
-                        }
-                    }
-                    // Emit a Provenance SEND event
-                    final long transmissionMillis = TimeUnit.NANOSECONDS.toMillis(System.nanoTime() - startNanos);
-
-                    session.getProvenanceReporter().send(flowFile, dbcpService.getConnectionURL(), transmissionMillis, true);
-                    session.transfer(flowFile, REL_SUCCESS);
-
-                } catch (final SQLException e) {
-
-                    if (e instanceof SQLNonTransientException) {
-                        getLogger().error("Failed to update Hive for {} due to {}; routing to failure", new Object[]{flowFile, e});
-                        session.transfer(flowFile, REL_FAILURE);
-                    } else {
-                        getLogger().error("Failed to update Hive for {} due to {}; it is possible that retrying the operation will succeed, so routing to retry", new Object[]{flowFile, e});
-                        flowFile = session.penalize(flowFile);
-                        session.transfer(flowFile, REL_RETRY);
-                    }
-
-                }
-            }
-        } catch (final SQLException sqle) {
-            // There was a problem getting the connection, yield and retry the flowfiles
-            getLogger().error("Failed to get Hive connection due to {}; it is possible that retrying the operation will succeed, so routing to retry", new Object[]{sqle});
-            session.transfer(flowFiles, REL_RETRY);
-            context.yield();
         }
     }
+
+    private InitConnection<FunctionContext, Connection> initConnection = (context, session, fc) -> {
+        final HiveDBCPService dbcpService = context.getProperty(HIVE_DBCP_SERVICE).asControllerService(HiveDBCPService.class);
+        final Connection connection = dbcpService.getConnection();
+        fc.connectionUrl = dbcpService.getConnectionURL();
+        return connection;
+    };
+
+    private FetchFlowFiles<FunctionContext> fetchFlowFiles = (context, session, functionContext, result) -> {
+        final int batchSize = context.getProperty(BATCH_SIZE).asInteger();
+        return session.get(batchSize);
+    };
+
+    private Put.PutFlowFile<FunctionContext, Connection> putFlowFile = (context, session, fc, conn, flowFile, result) -> {
+        final String script = getHiveQL(session, flowFile, fc.charset);
+        String regex = "(?<!\\\\)" + Pattern.quote(fc.statementDelimiter);
+
+        String[] hiveQLs = script.split(regex);
+
+        exceptionHandler.execute(fc, flowFile, input -> {
+            int loc = 1;
+            for (String hiveQL: hiveQLs) {
+                getLogger().debug("HiveQL: {}", new Object[]{hiveQL});
+
+                if (!StringUtils.isEmpty(hiveQL.trim())) {
+                    final PreparedStatement stmt = conn.prepareStatement(hiveQL.trim());
+
+                    // Get ParameterMetadata
+                    // Hive JDBC Doesn't support this yet:
+                    // ParameterMetaData pmd = stmt.getParameterMetaData();
+                    // int paramCount = pmd.getParameterCount();
+                    int paramCount = StringUtils.countMatches(hiveQL, "?");
+
+                    if (paramCount > 0) {
+                        loc = setParameters(loc, stmt, paramCount, flowFile.getAttributes());
+                    }
+
+                    // Execute the statement
+                    stmt.execute();
+                    fc.proceed();
+                }
+            }
+
+            // Emit a Provenance SEND event
+            final long transmissionMillis = TimeUnit.NANOSECONDS.toMillis(System.nanoTime() - fc.startNanos);
+
+            session.getProvenanceReporter().send(flowFile, fc.connectionUrl, transmissionMillis, true);
+            result.routeTo(flowFile, REL_SUCCESS);
+
+        }, onFlowFileError(context, session, result));
+
+    };
+
+    private OnError<FunctionContext, FlowFile> onFlowFileError(final ProcessContext context, final ProcessSession session, final RoutingResult result) {
+        OnError<FunctionContext, FlowFile> onFlowFileError = ExceptionHandler.createOnError(context, session, result, REL_FAILURE, REL_RETRY);
+        onFlowFileError = onFlowFileError.andThen((c, i, r, e) -> {
+            switch (r.destination()) {
+                case Failure:
+                    getLogger().error("Failed to update Hive for {} due to {}; routing to failure", new Object[] {i, e}, e);
+                    break;
+                case Retry:
+                    getLogger().error("Failed to update Hive for {} due to {}; it is possible that retrying the operation will succeed, so routing to retry",
+                            new Object[] {i, e}, e);
+                    break;
+            }
+        });
+        return RollbackOnFailure.createOnError(onFlowFileError);
+    }
+
+    @Override
+    public void onTrigger(ProcessContext context, ProcessSessionFactory sessionFactory) throws ProcessException {
+        final Boolean rollbackOnFailure = context.getProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE).asBoolean();
+        final Charset charset = Charset.forName(context.getProperty(CHARSET).getValue());
+        final String statementDelimiter = context.getProperty(STATEMENT_DELIMITER).getValue();
+        final FunctionContext functionContext = new FunctionContext(rollbackOnFailure, charset, statementDelimiter);
+        RollbackOnFailure.onTrigger(context, sessionFactory, functionContext, getLogger(), session -> process.onTrigger(context, session, functionContext));
+    }
 }

nifi-nar-bundles/nifi-hive-bundle/nifi-hive-processors/src/main/java/org/apache/nifi/processors/hive/PutHiveStreaming.java

@@ -17,7 +17,6 @@
 package org.apache.nifi.processors.hive;
 
 import com.google.common.util.concurrent.ThreadFactoryBuilder;
-import org.apache.avro.Schema;
 import org.apache.avro.file.CodecFactory;
 import org.apache.avro.file.DataFileConstants;
 import org.apache.avro.file.DataFileStream;
@@ -46,34 +45,38 @@ import org.apache.nifi.flowfile.FlowFile;
 import org.apache.nifi.hadoop.KerberosProperties;
 import org.apache.nifi.hadoop.SecurityUtil;
 import org.apache.nifi.logging.ComponentLog;
-import org.apache.nifi.processor.AbstractProcessor;
+import org.apache.nifi.processor.AbstractSessionFactoryProcessor;
 import org.apache.nifi.processor.ProcessContext;
 import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.ProcessSessionFactory;
 import org.apache.nifi.processor.ProcessorInitializationContext;
 import org.apache.nifi.processor.Relationship;
 import org.apache.nifi.processor.exception.ProcessException;
 import org.apache.nifi.processor.util.StandardValidators;
+import org.apache.nifi.processor.util.pattern.DiscontinuedException;
+import org.apache.nifi.processor.util.pattern.ErrorTypes;
+import org.apache.nifi.processor.util.pattern.ExceptionHandler;
+import org.apache.nifi.processor.util.pattern.RollbackOnFailure;
+import org.apache.nifi.processor.util.pattern.RoutingResult;
 import org.apache.nifi.util.hive.AuthenticationFailedException;
 import org.apache.nifi.util.hive.HiveConfigurator;
 import org.apache.nifi.util.hive.HiveOptions;
 import org.apache.nifi.util.hive.HiveUtils;
 import org.apache.nifi.util.hive.HiveWriter;
-import org.json.JSONException;
-import org.json.JSONObject;
 
 import java.io.File;
 import java.io.IOException;
 import java.nio.charset.StandardCharsets;
 import java.util.ArrayList;
-import java.util.Arrays;
 import java.util.Collections;
 import java.util.HashSet;
-import java.util.LinkedList;
 import java.util.List;
 import java.util.Map;
 import java.util.Set;
 import java.util.Timer;
 import java.util.TimerTask;
+import java.util.concurrent.ArrayBlockingQueue;
+import java.util.concurrent.BlockingQueue;
 import java.util.concurrent.ConcurrentHashMap;
 import java.util.concurrent.ExecutorService;
 import java.util.concurrent.Executors;
@@ -81,6 +84,8 @@ import java.util.concurrent.TimeUnit;
 import java.util.concurrent.atomic.AtomicBoolean;
 import java.util.concurrent.atomic.AtomicInteger;
 import java.util.concurrent.atomic.AtomicReference;
+import java.util.function.BiFunction;
+import java.util.function.Function;
 import java.util.regex.Pattern;
 
 /**
@@ -96,7 +101,7 @@ import java.util.regex.Pattern;
                 + "and 'failure' relationships, and contains the number of records from the incoming flow file written successfully and unsuccessfully, respectively.")
 })
 @RequiresInstanceClassLoading
-public class PutHiveStreaming extends AbstractProcessor {
+public class PutHiveStreaming extends AbstractSessionFactoryProcessor {
 
     // Attributes
     public static final String HIVE_STREAMING_RECORD_COUNT_ATTR = "hivestreaming.record.count";
@@ -231,21 +236,27 @@ public class PutHiveStreaming extends AbstractProcessor {
             .defaultValue("10000")
             .build();
 
+    public static final PropertyDescriptor ROLLBACK_ON_FAILURE = RollbackOnFailure.createRollbackOnFailureProperty(
+            "NOTE: When an error occurred after a Hive streaming transaction which is derived from the same input FlowFile is already committed," +
+                    " (i.e. a FlowFile contains more records than 'Records per Transaction' and a failure occurred at the 2nd transaction or later)" +
+                    " then the succeeded records will be transferred to 'success' relationship while the original input FlowFile stays in incoming queue." +
+                    " Duplicated records can be created for the succeeded ones when the same FlowFile is processed again.");
+
     // Relationships
     public static final Relationship REL_SUCCESS = new Relationship.Builder()
             .name("success")
-            .description("A FlowFile containing the JSON contents of a record is routed to this relationship after the record has been successfully transmitted to Hive.")
+            .description("A FlowFile containing Avro records routed to this relationship after the record has been successfully transmitted to Hive.")
             .build();
 
     public static final Relationship REL_FAILURE = new Relationship.Builder()
             .name("failure")
-            .description("A FlowFile containing the JSON contents of a record is routed to this relationship if the record could not be transmitted to Hive.")
+            .description("A FlowFile containing Avro records routed to this relationship if the record could not be transmitted to Hive.")
             .build();
 
     public static final Relationship REL_RETRY = new Relationship.Builder()
             .name("retry")
             .description("The incoming FlowFile is routed to this relationship if its records cannot be transmitted to Hive. Note that "
-                    + "some records may have been processed successfully, they will be routed (as JSON flow files) to the success relationship. "
+                    + "some records may have been processed successfully, they will be routed (as Avro flow files) to the success relationship. "
                     + "The combination of the retry, success, and failure relationships indicate how many records succeeded and/or failed. This "
                     + "can be used to provide a retry capability since full rollback is not possible.")
             .build();
@@ -283,6 +294,7 @@ public class PutHiveStreaming extends AbstractProcessor {
         props.add(HEARTBEAT_INTERVAL);
         props.add(TXNS_PER_BATCH);
         props.add(RECORDS_PER_TXN);
+        props.add(ROLLBACK_ON_FAILURE);
 
         kerberosConfigFile = context.getKerberosConfigurationFile();
         kerberosProperties = new KerberosProperties(kerberosConfigFile);
@@ -364,8 +376,213 @@ public class PutHiveStreaming extends AbstractProcessor {
         setupHeartBeatTimer();
     }
 
+    private static class FunctionContext extends RollbackOnFailure {
+
+        private FlowFile inputFlowFile;
+        private AtomicReference<FlowFile> successFlowFile;
+        private AtomicReference<FlowFile> failureFlowFile;
+        private final DataFileWriter<GenericRecord> successAvroWriter = new DataFileWriter<>(new GenericDatumWriter<GenericRecord>());
+        private final DataFileWriter<GenericRecord> failureAvroWriter = new DataFileWriter<>(new GenericDatumWriter<GenericRecord>());
+
+        private final AtomicInteger recordCount = new AtomicInteger(0);
+        private final AtomicInteger successfulRecordCount = new AtomicInteger(0);
+        private final AtomicInteger failedRecordCount = new AtomicInteger(0);
+
+        private volatile ExecutorService appendRecordThreadPool;
+        private volatile AtomicBoolean closed = new AtomicBoolean(false);
+        private final BlockingQueue<List<HiveStreamingRecord>> successRecordQueue = new ArrayBlockingQueue<>(100);
+        private final BlockingQueue<List<HiveStreamingRecord>> failureRecordQueue = new ArrayBlockingQueue<>(100);
+
+        private final ComponentLog logger;
+
+        /**
+         * It's possible that multiple Hive streaming transactions are committed within a single onTrigger.
+         * PutHiveStreaming onTrigger is not 'transactional' in a sense of RollbackOnFailure.
+         * Once a Hive streaming transaction is committed, processor session will not be rolled back.
+         * @param rollbackOnFailure whether process session should be rolled back if failed
+         */
+        private FunctionContext(boolean rollbackOnFailure, ComponentLog logger) {
+            super(rollbackOnFailure, false);
+            this.logger = logger;
+        }
+
+        private void setFlowFiles(FlowFile inputFlowFile, FlowFile successFlowFile, FlowFile failureFlowFile) {
+            this.inputFlowFile = inputFlowFile;
+            this.successFlowFile = new AtomicReference<>(successFlowFile);
+            this.failureFlowFile = new AtomicReference<>(failureFlowFile);
+        }
+
+        private void initAvroWriter(ProcessSession session, String codec, DataFileStream<GenericRecord> reader,
+                                    DataFileWriter<GenericRecord> writer, AtomicReference<FlowFile> flowFileRef,
+                                    BlockingQueue<List<HiveStreamingRecord>> queue, Function<Integer, Boolean> isCompleted) {
+
+            writer.setCodec(CodecFactory.fromString(codec));
+            // Transfer metadata (this is a subset of the incoming file)
+            for (String metaKey : reader.getMetaKeys()) {
+                if (!RESERVED_METADATA.contains(metaKey)) {
+                    writer.setMeta(metaKey, reader.getMeta(metaKey));
+                }
+            }
+
+            appendRecordThreadPool.submit(() -> {
+                flowFileRef.set(session.append(flowFileRef.get(), (out) -> {
+                    // Create writer so that records can be appended.
+                    writer.create(reader.getSchema(), out);
+
+                    try {
+                        int writtenCount = 0;
+                        while (true) {
+
+                            if (closed.get() && isCompleted.apply(writtenCount)) {
+                                break;
+                            }
+
+                            final List<HiveStreamingRecord> hRecords = queue.poll(100, TimeUnit.MILLISECONDS);
+                            if (hRecords != null) {
+                                try {
+                                    for (HiveStreamingRecord hRecord : hRecords) {
+                                        writer.append(hRecord.getRecord());
+                                        writtenCount++;
+                                    }
+                                } catch (IOException ioe) {
+                                    // The records were put to Hive Streaming successfully, but there was an error while writing the
+                                    // Avro records to the flow file. Log as an error and move on.
+                                    logger.error("Error writing Avro records (which were sent successfully to Hive Streaming) to the flow file, " + ioe, ioe);
+                                }
+                            }
+                        }
+                        writer.flush();
+                    } catch (InterruptedException e) {
+                        logger.warn("Append record thread is interrupted, " + e, e);
+                    }
+
+                }));
+            });
+        }
+
+        private void initAvroWriters(ProcessSession session, String codec, DataFileStream<GenericRecord> reader) {
+            appendRecordThreadPool = Executors.newFixedThreadPool(2);
+            initAvroWriter(session, codec, reader, successAvroWriter, successFlowFile, successRecordQueue, w -> w == successfulRecordCount.get());
+            initAvroWriter(session, codec, reader, failureAvroWriter, failureFlowFile, failureRecordQueue, w -> w == failedRecordCount.get());
+
+            // No new task.
+            appendRecordThreadPool.shutdown();
+        }
+
+        private void appendRecordsToSuccess(List<HiveStreamingRecord> records) {
+            appendRecordsToFlowFile(records, successRecordQueue);
+            successfulRecordCount.addAndGet(records.size());
+        }
+
+        private void appendRecordsToFailure(List<HiveStreamingRecord> records) {
+            appendRecordsToFlowFile(records, failureRecordQueue);
+            failedRecordCount.addAndGet(records.size());
+        }
+
+        private void appendRecordsToFlowFile(List<HiveStreamingRecord> records, BlockingQueue<List<HiveStreamingRecord>> queue) {
+            if (!queue.add(records)) {
+                throw new ProcessException(String.format("Failed to append %d records due to insufficient internal queue capacity.", records.size()));
+            }
+        }
+
+        private void transferFlowFiles(ProcessSession session, RoutingResult result, String transitUri) {
+
+            closeAvroWriters();
+
+            if (successfulRecordCount.get() > 0) {
+                // Transfer the flow file with successful records
+                successFlowFile.set(
+                        session.putAttribute(successFlowFile.get(), HIVE_STREAMING_RECORD_COUNT_ATTR, Integer.toString(successfulRecordCount.get())));
+                session.getProvenanceReporter().send(successFlowFile.get(), transitUri);
+                result.routeTo(successFlowFile.get(), REL_SUCCESS);
+            } else {
+                session.remove(successFlowFile.get());
+            }
+
+            if (failedRecordCount.get() > 0) {
+                // There were some failed records, so transfer that flow file to failure
+                failureFlowFile.set(
+                        session.putAttribute(failureFlowFile.get(), HIVE_STREAMING_RECORD_COUNT_ATTR, Integer.toString(failedRecordCount.get())));
+                result.routeTo(failureFlowFile.get(), REL_FAILURE);
+            } else {
+                session.remove(failureFlowFile.get());
+            }
+
+            result.getRoutedFlowFiles().forEach((relationship, flowFiles) -> {
+                session.transfer(flowFiles, relationship);
+            });
+        }
+
+        private void closeAvroWriters() {
+            closed.set(true);
+            if (appendRecordThreadPool != null) {
+                // Having null thread pool means the input FlowFile was not processed at all, due to illegal format.
+                try {
+                    if (!appendRecordThreadPool.awaitTermination(10, TimeUnit.SECONDS)) {
+                        logger.warn("Waiting for Avro records being appended into output FlowFiles has been timeout.");
+                    }
+                } catch (InterruptedException e) {
+                    logger.warn("Waiting for Avro records being appended into output FlowFiles has been interrupted.");
+                }
+            }
+        }
+    }
+
+    private static class ShouldRetryException extends RuntimeException {
+        private ShouldRetryException(String message, Throwable cause) {
+            super(message, cause);
+        }
+    }
+
+    private ExceptionHandler.OnError<FunctionContext, List<HiveStreamingRecord>> onHiveRecordsError(ProcessContext context, ProcessSession session) {
+        return RollbackOnFailure.createOnError((fc, input, res, e) -> {
+
+            if (res.penalty() == ErrorTypes.Penalty.Yield) {
+                context.yield();
+            }
+
+            switch (res.destination()) {
+                case Failure:
+                    // Add the failed record to the failure flow file
+                    getLogger().error(String.format("Error writing %s to Hive Streaming transaction due to %s", input, e), e);
+                    fc.appendRecordsToFailure(input);
+                    break;
+
+                case Retry:
+                    // If we can't connect to the endpoint, exit the loop and let the outer exception handler route the original flow file to retry
+                    abortAndCloseWriters();
+                    throw new ShouldRetryException("Hive Streaming connect/write error, flow file will be penalized and routed to retry. " + e, e);
+
+                case Self:
+                    abortAndCloseWriters();
+                    break;
+
+                default:
+                    abortAndCloseWriters();
+                    if (e instanceof ProcessException) {
+                        throw (ProcessException) e;
+                    } else {
+                        throw new ProcessException(String.format("Error writing %s to Hive Streaming transaction due to %s", input, e), e);
+                    }
+            }
+        });
+    }
+
+    private ExceptionHandler.OnError<FunctionContext, HiveStreamingRecord> onHiveRecordError(ProcessContext context, ProcessSession session) {
+        return (fc, input, res, e) -> onHiveRecordsError(context, session).apply(fc, Collections.singletonList(input), res, e);
+    }
+
+    private ExceptionHandler.OnError<FunctionContext, GenericRecord> onRecordError(ProcessContext context, ProcessSession session) {
+        return (fc, input, res, e) -> onHiveRecordError(context, session).apply(fc, new HiveStreamingRecord(null, input), res, e);
+    }
+
     @Override
-    public void onTrigger(ProcessContext context, ProcessSession session) throws ProcessException {
+    public void onTrigger(ProcessContext context, ProcessSessionFactory sessionFactory) throws ProcessException {
+        final FunctionContext functionContext = new FunctionContext(context.getProperty(ROLLBACK_ON_FAILURE).asBoolean(), getLogger());
+        RollbackOnFailure.onTrigger(context, sessionFactory, functionContext, getLogger(), session -> onTrigger(context, session, functionContext));
+    }
+
+    private void onTrigger(ProcessContext context, ProcessSession session, FunctionContext functionContext) throws ProcessException {
         FlowFile flowFile = session.get();
         if (flowFile == null) {
             return;
@@ -390,22 +607,58 @@ public class PutHiveStreaming extends AbstractProcessor {
             }
         }
 
-        final AtomicInteger recordCount = new AtomicInteger(0);
+        final AtomicReference<List<HiveStreamingRecord>> successfulRecords = new AtomicReference<>();
-        final AtomicInteger successfulRecordCount = new AtomicInteger(0);
+        successfulRecords.set(new ArrayList<>());
-        List<HiveStreamingRecord> successfulRecords = new LinkedList<>();
         final FlowFile inputFlowFile = flowFile;
-        final AtomicBoolean processingFailure = new AtomicBoolean(false);
+
+        final RoutingResult result = new RoutingResult();
+        final ExceptionHandler<FunctionContext> exceptionHandler = new ExceptionHandler<>();
+        exceptionHandler.mapException(s -> {
+            try {
+                if (s == null) {
+                    return ErrorTypes.PersistentFailure;
+                }
+                throw s;
+
+            } catch (IllegalArgumentException
+                    | HiveWriter.WriteFailure
+                    | SerializationError inputError) {
+
+                return ErrorTypes.InvalidInput;
+
+            } catch (HiveWriter.CommitFailure
+                    | HiveWriter.TxnBatchFailure
+                    | HiveWriter.TxnFailure writerTxError) {
+
+                return ErrorTypes.TemporalInputFailure;
+
+            } catch (ConnectionError
+                    | HiveWriter.ConnectFailure connectionError) {
+                // Can't connect to Hive endpoint.
+                log.error("Error connecting to Hive endpoint: table {} at {}",
+                        new Object[]{options.getTableName(), options.getMetaStoreURI()});
+
+                return ErrorTypes.TemporalFailure;
+
+            } catch (IOException
+                    | InterruptedException tempError) {
+                return ErrorTypes.TemporalFailure;
+
+            } catch (Exception t) {
+                return ErrorTypes.UnknownFailure;
+            }
+        });
+        final BiFunction<FunctionContext, ErrorTypes, ErrorTypes.Result> adjustError = RollbackOnFailure.createAdjustError(getLogger());
+        exceptionHandler.adjustError(adjustError);
 
         // Create output flow files and their Avro writers
-        AtomicReference<FlowFile> successFlowFile = new AtomicReference<>(session.create(inputFlowFile));
+        functionContext.setFlowFiles(inputFlowFile, session.create(inputFlowFile), session.create(inputFlowFile));
-        final DataFileWriter<GenericRecord> successAvroWriter = new DataFileWriter<>(new GenericDatumWriter<GenericRecord>());
-        AtomicReference<FlowFile> failureFlowFile = new AtomicReference<>(session.create(inputFlowFile));
-        final DataFileWriter<GenericRecord> failureAvroWriter = new DataFileWriter<>(new GenericDatumWriter<GenericRecord>());
 
         try {
             session.read(inputFlowFile, in -> {
 
                 try (final DataFileStream<GenericRecord> reader = new DataFileStream<>(in, new GenericDatumReader<GenericRecord>())) {
+
                     GenericRecord currRecord = null;
 
                     // Copy codec and schema information to all writers
@@ -413,239 +666,133 @@ public class PutHiveStreaming extends AbstractProcessor {
                             ? DataFileConstants.NULL_CODEC
                             : reader.getMetaString(DataFileConstants.CODEC);
 
-                    Arrays.asList(successAvroWriter, failureAvroWriter)
+                    functionContext.initAvroWriters(session, codec, reader);
-                            .forEach((writer) -> {
+
-                                writer.setCodec(CodecFactory.fromString(codec));
+                    Runnable flushSuccessfulRecords = () -> {
-                                // Transfer metadata (this is a subset of the incoming file)
+                        // Now send the records to the successful FlowFile and update the success count
-                                for (String metaKey : reader.getMetaKeys()) {
+                        functionContext.appendRecordsToSuccess(successfulRecords.get());
-                                    if (!RESERVED_METADATA.contains(metaKey)) {
+                        // Clear the list of successful records, we'll use it at the end when we flush whatever records are left
-                                        writer.setMeta(metaKey, reader.getMeta(metaKey));
+                        successfulRecords.set(new ArrayList<>());
-                                    }
+                    };
-                                }
-                            });
 
                     while (reader.hasNext()) {
-                        currRecord = reader.next(currRecord);
+                        // We can NOT reuse currRecord here, because currRecord is accumulated in successful records.
-                        recordCount.incrementAndGet();
+                        // If we use the same GenericRecord instance, every record ends up having the same contents.
+                        // To avoid this, we need to create a brand new GenericRecord instance here each time.
+                        currRecord = reader.next();
+                        functionContext.recordCount.incrementAndGet();
 
                         // Extract the partition values (they must be put separately into the Hive Streaming API)
                         List<String> partitionValues = new ArrayList<>();
 
-                        try {
+                        if (!exceptionHandler.execute(functionContext, currRecord, input -> {
                             for (String partition : partitionColumnList) {
-                                Object partitionValue = currRecord.get(partition);
+                                Object partitionValue = input.get(partition);
                                 if (partitionValue == null) {
-                                    throw new IOException("Partition column '" + partition + "' not found in Avro record");
+                                    throw new IllegalArgumentException("Partition column '" + partition + "' not found in Avro record");
                                 }
                                 partitionValues.add(partitionValue.toString());
                             }
-                        } catch (IOException ioe) {
+                        }, onRecordError(context, session))) {
-                            // Add the failed record to the failure flow file
-                            log.error("Error writing record to Hive Streaming transaction", ioe);
-                            appendRecordsToFlowFile(session, Collections.singletonList(new HiveStreamingRecord(null, currRecord)),
-                                    failureFlowFile, failureAvroWriter, reader);
                             continue;
                         }
 
-                        List<Schema.Field> fields = currRecord.getSchema().getFields();
+                        final HiveStreamingRecord record = new HiveStreamingRecord(partitionValues, currRecord);
-                        if (fields != null) {
+                        final AtomicReference<HiveWriter> hiveWriterRef = new AtomicReference<>();
-                            JSONObject obj = new JSONObject();
+
-                            try {
+                        // Write record to Hive streaming
-                                for (Schema.Field field : fields) {
+                        if (!exceptionHandler.execute(functionContext, record, input -> {
-                                    String fieldName = field.name();
+
-                                    // Skip fields that are partition columns, we extracted those values above to create an EndPoint
+                            final HiveEndPoint endPoint = makeHiveEndPoint(record.getPartitionValues(), options);
-                                    if (!partitionColumnList.contains(fieldName)) {
+                            final HiveWriter hiveWriter = getOrCreateWriter(endPoint);
-                                        Object value = currRecord.get(fieldName);
+                            hiveWriterRef.set(hiveWriter);
+
+                            hiveWriter.write(record.getRecord().toString().getBytes(StandardCharsets.UTF_8));
+                            successfulRecords.get().add(record);
+
+                        }, onHiveRecordError(context, session))) {
+                            continue;
+                        }
+
+                        // If we've reached the records-per-transaction limit, flush the Hive Writer and update the Avro Writer for successful records
+                        final HiveWriter hiveWriter = hiveWriterRef.get();
+                        if (hiveWriter.getTotalRecords() >= recordsPerTxn) {
+                            exceptionHandler.execute(functionContext, successfulRecords.get(), input -> {
+
+                                hiveWriter.flush(true);
+                                // Proceed function context. Process session can't be rollback anymore.
+                                functionContext.proceed();
+
+                                // Now send the records to the success relationship and update the success count
+                                flushSuccessfulRecords.run();
+
+                            }, onHiveRecordsError(context, session).andThen((fc, input, res, commitException) -> {
+                                // Reset hiveWriter for succeeding records.
+                                switch (res.destination()) {
+                                    case Retry:
+                                    case Failure:
                                         try {
-                                            obj.put(fieldName, value);
+                                            // Abort current tx and move to next.
-                                        } catch (JSONException je) {
+                                            hiveWriter.abort();
-                                            throw new IOException(je);
+                                        } catch (Exception e) {
+                                            // Can't even abort properly, throw a process exception
+                                            throw new ProcessException(e);
                                         }
-                                    }
                                 }
-                            } catch (IOException ioe) {
+                            }));
-                                // This really shouldn't happen since we are iterating over the schema fields, but just in case,
-                                // add the failed record to the failure flow file.
-                                log.error("Error writing record to Hive Streaming transaction", ioe);
-                                appendRecordsToFlowFile(session, Collections.singletonList(new HiveStreamingRecord(null, currRecord)),
-                                        failureFlowFile, failureAvroWriter, reader);
-                                continue;
-                            }
-                            final HiveStreamingRecord record = new HiveStreamingRecord(partitionValues, currRecord);
-                            HiveEndPoint endPoint = null;
-                            HiveWriter hiveWriter = null;
-                            try {
-                                endPoint = makeHiveEndPoint(record.getPartitionValues(), options);
-                                hiveWriter = getOrCreateWriter(endPoint);
-                            } catch (ConnectionError
-                                    | HiveWriter.ConnectFailure
-                                    | InterruptedException connectionError) {
-                                // Can't connect to Hive endpoint.
-                                log.error("Error connecting to Hive endpoint: table {} at {}",
-                                        new Object[]{options.getTableName(), options.getMetaStoreURI()});
-                                // If we can't connect to the endpoint, exit the loop and let the outer exception handler route the original flow file to retry
-                                abortAndCloseWriters();
-                                throw new ProcessException(connectionError);
-                            }
-                            try {
-                                try {
-                                    hiveWriter.write(record.getRecord().toString().getBytes(StandardCharsets.UTF_8));
-                                    successfulRecords.add(record);
-                                } catch (InterruptedException | HiveWriter.WriteFailure wf) {
-                                    // Add the failed record to the failure flow file
-                                    log.error("Error writing record to Hive Streaming transaction", wf);
-                                    appendRecordsToFlowFile(session, Collections.singletonList(record), failureFlowFile, failureAvroWriter, reader);
-                                }
-
-                                // If we've reached the records-per-transaction limit, flush the Hive Writer and update the Avro Writer for successful records
-                                if (hiveWriter.getTotalRecords() >= recordsPerTxn) {
-                                    hiveWriter.flush(true);
-                                    // Now send the records to the success relationship and update the success count
-                                    try {
-                                        appendRecordsToFlowFile(session, successfulRecords, successFlowFile, successAvroWriter, reader);
-                                        successfulRecordCount.accumulateAndGet(successfulRecords.size(), (current, incr) -> current + incr);
-
-                                        // Clear the list of successful records, we'll use it at the end when we flush whatever records are left
-                                        successfulRecords.clear();
-
-                                    } catch (IOException ioe) {
-                                        // The records were put to Hive Streaming successfully, but there was an error while writing the
-                                        // Avro records to the flow file. Log as an error and move on.
-                                        getLogger().error("Error writing Avro records (which were sent successfully to Hive Streaming) to the flow file", ioe);
-                                    }
-                                }
-
-                            } catch (InterruptedException
-                                    | HiveWriter.CommitFailure
-                                    | HiveWriter.TxnBatchFailure
-                                    | HiveWriter.TxnFailure
-                                    | SerializationError writeException) {
-
-                                log.error("Error writing record to Hive Streaming transaction", writeException);
-                                // Add the failed record to the failure flow file
-                                appendRecordsToFlowFile(session, Collections.singletonList(record), failureFlowFile, failureAvroWriter, reader);
-
-                                if (!(writeException instanceof SerializationError)) {
-                                    try {
-                                        hiveWriter.abort();
-                                    } catch (Exception e) {
-                                        // Can't even abort properly, throw a process exception
-                                        throw new ProcessException(e);
-                                    }
-                                }
-                            }
                         }
                     }
-                    try {
+
+                    exceptionHandler.execute(functionContext, successfulRecords.get(), input -> {
                         // Finish any transactions
                         flushAllWriters(true);
                         closeAllWriters();
 
                         // Now send any remaining records to the success relationship and update the count
-                        appendRecordsToFlowFile(session, successfulRecords, successFlowFile, successAvroWriter, reader);
+                        flushSuccessfulRecords.run();
-                        successfulRecordCount.accumulateAndGet(successfulRecords.size(), (current, incr) -> current + incr);
-                        successfulRecords.clear();
 
-                    } catch (HiveWriter.CommitFailure
+                        // Append successfulRecords on failure.
-                            | HiveWriter.TxnBatchFailure
+                    }, onHiveRecordsError(context, session));
-                            | HiveWriter.TxnFailure
-                            | InterruptedException e) {
 
-                        // If any records are in the successfulRecords list but ended up here, then they actually weren't transferred successfully, so
-                        // route them to failure instead
-                        appendRecordsToFlowFile(session, successfulRecords, failureFlowFile, failureAvroWriter, reader);
-                    }
                 } catch (IOException ioe) {
                     // The Avro file is invalid (or may not be an Avro file at all), send it to failure
-                    log.error("The incoming flow file can not be read as an Avro file, routing to failure", ioe);
+                    final ErrorTypes.Result adjusted = adjustError.apply(functionContext, ErrorTypes.InvalidInput);
-                    processingFailure.set(true);
+                    final String msg = "The incoming flow file can not be read as an Avro file";
+                    switch (adjusted.destination()) {
+                        case Failure:
+                            log.error(msg, ioe);
+                            result.routeTo(inputFlowFile, REL_FAILURE);
+                            break;
+                        case ProcessException:
+                            throw new ProcessException(msg, ioe);
+
+                    }
                 }
             });
 
-            if (recordCount.get() > 0) {
-                if (successfulRecordCount.get() > 0) {
-                    // Transfer the flow file with successful records
-                    successFlowFile.set(
-                            session.putAttribute(successFlowFile.get(), HIVE_STREAMING_RECORD_COUNT_ATTR, Integer.toString(recordCount.get())));
-                    session.getProvenanceReporter().send(successFlowFile.get(), options.getMetaStoreURI());
-                    session.transfer(successFlowFile.get(), REL_SUCCESS);
-                } else {
-                    session.remove(successFlowFile.get());
-                }
-
-                if (recordCount.get() != successfulRecordCount.get()) {
-                    // There were some failed records, so transfer that flow file to failure
-                    failureFlowFile.set(
-                            session.putAttribute(failureFlowFile.get(), HIVE_STREAMING_RECORD_COUNT_ATTR,
-                                    Integer.toString(recordCount.get() - successfulRecordCount.get())));
-                    session.transfer(failureFlowFile.get(), REL_FAILURE);
-                } else {
-                    session.remove(failureFlowFile.get());
-                }
-            } else {
-                // No records were processed, so remove the output flow files
-                session.remove(successFlowFile.get());
-                session.remove(failureFlowFile.get());
-            }
-            successFlowFile.set(null);
-            failureFlowFile.set(null);
-
             // If we got here, we've processed the outgoing flow files correctly, so remove the incoming one if necessary
-            if (processingFailure.get()) {
+            if (result.getRoutedFlowFiles().values().stream().noneMatch(routed -> routed.contains(inputFlowFile))) {
-                session.transfer(inputFlowFile, REL_FAILURE);
+                session.remove(inputFlowFile);
-            } else {
-                session.remove(flowFile);
             }
 
-        } catch (ProcessException pe) {
+        } catch (DiscontinuedException e) {
-            abortAndCloseWriters();
+            // The input FlowFile processing is discontinued. Keep it in the input queue.
-            Throwable t = pe.getCause();
+            getLogger().warn("Discontinued processing for {} due to {}", new Object[]{flowFile, e}, e);
-            if (t != null) {
+            result.routeTo(flowFile, Relationship.SELF);
-                if (t instanceof ConnectionError
+
-                        || t instanceof HiveWriter.ConnectFailure
+        } catch (ShouldRetryException e) {
-                        || t instanceof HiveWriter.CommitFailure
+            // This exception is already a result of adjusting an error, so simply transfer the FlowFile to retry.
-                        || t instanceof HiveWriter.TxnBatchFailure
+            getLogger().error(e.getMessage(), e);
-                        || t instanceof HiveWriter.TxnFailure
+            flowFile = session.penalize(flowFile);
-                        || t instanceof InterruptedException) {
+            result.routeTo(flowFile, REL_RETRY);
-                    log.error("Hive Streaming connect/write error, flow file will be penalized and routed to retry", t);
+
-                    flowFile = session.penalize(flowFile);
-                    session.transfer(flowFile, REL_RETRY);
-                    // Remove the ones we created
-                    if (successFlowFile.get() != null) {
-                        session.remove(successFlowFile.get());
-                    }
-                    if (failureFlowFile.get() != null) {
-                        session.remove(failureFlowFile.get());
-                    }
-                } else {
-                    throw pe;
-                }
-            } else {
-                throw pe;
-            }
         } finally {
+            functionContext.transferFlowFiles(session, result, options.getMetaStoreURI());
             // Restore original class loader, might not be necessary but is good practice since the processor task changed it
             Thread.currentThread().setContextClassLoader(originalClassloader);
         }
     }
 
-    private void appendRecordsToFlowFile(ProcessSession session,
-                                         List<HiveStreamingRecord> records,
-                                         AtomicReference<FlowFile> appendFlowFile,
-                                         DataFileWriter<GenericRecord> avroWriter,
-                                         DataFileStream<GenericRecord> reader) throws IOException {
-
-        appendFlowFile.set(session.append(appendFlowFile.get(), (out) -> {
-
-            try (DataFileWriter<GenericRecord> writer = avroWriter.create(reader.getSchema(), out)) {
-                for (HiveStreamingRecord sRecord : records) {
-                    writer.append(sRecord.getRecord());
-                }
-                writer.flush();
-            }
-        }));
-    }
-
     @OnStopped
     public void cleanup() {
         ComponentLog log = getLogger();

nifi-nar-bundles/nifi-hive-bundle/nifi-hive-processors/src/main/java/org/apache/nifi/processors/hive/SelectHiveQL.java

@@ -50,11 +50,13 @@ import org.apache.nifi.flowfile.attributes.CoreAttributes;
 import org.apache.nifi.logging.ComponentLog;
 import org.apache.nifi.processor.ProcessContext;
 import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.ProcessSessionFactory;
 import org.apache.nifi.processor.Relationship;
 import org.apache.nifi.processor.exception.ProcessException;
 import org.apache.nifi.processor.io.InputStreamCallback;
 import org.apache.nifi.processor.io.OutputStreamCallback;
 import org.apache.nifi.processor.util.StandardValidators;
+import org.apache.nifi.processor.util.pattern.PartialFunctions;
 import org.apache.nifi.util.StopWatch;
 import org.apache.nifi.util.hive.CsvOutputOptions;
 import org.apache.nifi.util.hive.HiveJdbcCommon;
@@ -209,7 +211,11 @@ public class SelectHiveQL extends AbstractHiveQLProcessor {
     }
 
     @Override
-    public void onTrigger(final ProcessContext context, final ProcessSession session) throws ProcessException {
+    public void onTrigger(ProcessContext context, ProcessSessionFactory sessionFactory) throws ProcessException {
+        PartialFunctions.onTrigger(context, sessionFactory, getLogger(), session -> onTrigger(context, session));
+    }
+
+    private void onTrigger(final ProcessContext context, final ProcessSession session) throws ProcessException {
         final FlowFile fileToProcess = (context.hasIncomingConnection()? session.get():null);
         FlowFile flowfile = null;
 

nifi-nar-bundles/nifi-hive-bundle/nifi-hive-processors/src/test/java/org/apache/nifi/processors/hive/TestPutHiveQL.java

@@ -19,6 +19,7 @@ import org.apache.nifi.controller.AbstractControllerService;
 import org.apache.nifi.dbcp.DBCPService;
 import org.apache.nifi.dbcp.hive.HiveDBCPService;
 import org.apache.nifi.processor.exception.ProcessException;
+import org.apache.nifi.processor.util.pattern.RollbackOnFailure;
 import org.apache.nifi.reporting.InitializationException;
 import org.apache.nifi.util.TestRunner;
 import org.apache.nifi.util.TestRunners;
@@ -42,6 +43,7 @@ import java.util.Map;
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertTrue;
+import static org.junit.Assert.fail;
 
 public class TestPutHiveQL {
     private static final String createPersons = "CREATE TABLE PERSONS (id integer primary key, name varchar(100), code integer)";
@@ -128,6 +130,91 @@ public class TestPutHiveQL {
         runner.assertTransferCount(PutHiveQL.REL_SUCCESS, 3);
     }
 
+    @Test
+    public void testFailInMiddleWithBadStatementRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutHiveQL.class);
+        final File tempDir = folder.getRoot();
+        final File dbDir = new File(tempDir, "db");
+        final DBCPService service = new MockDBCPService(dbDir.getAbsolutePath());
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                stmt.executeUpdate(createPersonsAutoId);
+            }
+        }
+
+        runner.setProperty(PutHiveQL.HIVE_DBCP_SERVICE, "dbcp");
+        runner.enqueue("INSERT INTO PERSONS (NAME, CODE) VALUES ('Mark', 84)".getBytes());
+        runner.enqueue("INSERT INTO PERSONS".getBytes()); // intentionally wrong syntax
+        runner.enqueue("INSERT INTO PERSONS (NAME, CODE) VALUES ('Tom', 3)".getBytes());
+        runner.enqueue("INSERT INTO PERSONS (NAME, CODE) VALUES ('Harry', 44)".getBytes());
+        runner.run();
+
+        // The 1st one should be routed to success, others should stay in queue.
+        assertEquals(3, runner.getQueueSize().getObjectCount());
+        runner.assertTransferCount(PutHiveQL.REL_FAILURE, 0);
+        runner.assertTransferCount(PutHiveQL.REL_SUCCESS, 1);
+    }
+
+    @Test
+    public void testFailAtBeginning() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutHiveQL.class);
+        final File tempDir = folder.getRoot();
+        final File dbDir = new File(tempDir, "db");
+        final DBCPService service = new MockDBCPService(dbDir.getAbsolutePath());
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                stmt.executeUpdate(createPersonsAutoId);
+            }
+        }
+
+        runner.setProperty(PutHiveQL.HIVE_DBCP_SERVICE, "dbcp");
+        runner.enqueue("INSERT INTO PERSONS".getBytes()); // intentionally wrong syntax
+        runner.enqueue("INSERT INTO PERSONS (NAME, CODE) VALUES ('Tom', 3)".getBytes());
+        runner.enqueue("INSERT INTO PERSONS (NAME, CODE) VALUES ('Harry', 44)".getBytes());
+        runner.run();
+
+        runner.assertTransferCount(PutHiveQL.REL_FAILURE, 1);
+        runner.assertTransferCount(PutHiveQL.REL_SUCCESS, 2);
+    }
+
+    @Test
+    public void testFailAtBeginningRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutHiveQL.class);
+        final File tempDir = folder.getRoot();
+        final File dbDir = new File(tempDir, "db");
+        final DBCPService service = new MockDBCPService(dbDir.getAbsolutePath());
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                stmt.executeUpdate(createPersonsAutoId);
+            }
+        }
+
+        runner.setProperty(PutHiveQL.HIVE_DBCP_SERVICE, "dbcp");
+        runner.enqueue("INSERT INTO PERSONS".getBytes()); // intentionally wrong syntax
+        runner.enqueue("INSERT INTO PERSONS (NAME, CODE) VALUES ('Tom', 3)".getBytes());
+        runner.enqueue("INSERT INTO PERSONS (NAME, CODE) VALUES ('Harry', 44)".getBytes());
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        assertEquals(3, runner.getQueueSize().getObjectCount());
+        runner.assertTransferCount(PutHiveQL.REL_FAILURE, 0);
+        runner.assertTransferCount(PutHiveQL.REL_SUCCESS, 0);
+    }
 
     @Test
     public void testFailInMiddleWithBadParameterType() throws InitializationException, ProcessException, SQLException, IOException {
@@ -189,7 +276,56 @@ public class TestPutHiveQL {
 
         final Map<String, String> badAttributes = new HashMap<>();
         badAttributes.put("hiveql.args.1.type", String.valueOf(Types.INTEGER));
-        badAttributes.put("hiveql.args.1.value", "9999");
+        badAttributes.put("hiveql.args.1.value", "101"); // Constraint violation, up to 100
+
+        final byte[] data = "INSERT INTO PERSONS (NAME, CODE) VALUES ('Mark', ?)".getBytes();
+        runner.enqueue(data, goodAttributes);
+        runner.enqueue(data, badAttributes);
+        runner.enqueue(data, goodAttributes);
+        runner.enqueue(data, goodAttributes);
+        runner.run();
+
+        runner.assertTransferCount(PutHiveQL.REL_SUCCESS, 3);
+        runner.assertTransferCount(PutHiveQL.REL_FAILURE, 1);
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                final ResultSet rs = stmt.executeQuery("SELECT * FROM PERSONS");
+                assertTrue(rs.next());
+                assertEquals(1, rs.getInt(1));
+                assertEquals("Mark", rs.getString(2));
+                assertEquals(84, rs.getInt(3));
+                assertTrue(rs.next());
+                assertTrue(rs.next());
+                assertFalse(rs.next());
+            }
+        }
+    }
+
+    @Test
+    public void testFailInMiddleWithBadNumberFormat() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutHiveQL.class);
+        final File tempDir = folder.getRoot();
+        final File dbDir = new File(tempDir, "db");
+        final DBCPService service = new MockDBCPService(dbDir.getAbsolutePath());
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                stmt.executeUpdate(createPersonsAutoId);
+            }
+        }
+
+        runner.setProperty(PutHiveQL.HIVE_DBCP_SERVICE, "dbcp");
+
+        final Map<String, String> goodAttributes = new HashMap<>();
+        goodAttributes.put("hiveql.args.1.type", String.valueOf(Types.INTEGER));
+        goodAttributes.put("hiveql.args.1.value", "84");
+
+        final Map<String, String> badAttributes = new HashMap<>();
+        badAttributes.put("hiveql.args.1.type", String.valueOf(Types.INTEGER));
+        badAttributes.put("hiveql.args.1.value", "NOT_NUMBER");
 
         final byte[] data = "INSERT INTO PERSONS (NAME, CODE) VALUES ('Mark', ?)".getBytes();
         runner.enqueue(data, goodAttributes);
@@ -540,6 +676,44 @@ public class TestPutHiveQL {
         runner.assertAllFlowFilesTransferred(PutHiveQL.REL_RETRY, 1);
     }
 
+    @Test
+    public void testRetryableFailureRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutHiveQL.class);
+        final DBCPService service = new SQLExceptionService(null);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+
+        runner.setProperty(PutHiveQL.HIVE_DBCP_SERVICE, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        final String sql = "INSERT INTO PERSONS (ID, NAME, CODE) VALUES (?, ?, ?); " +
+                "UPDATE PERSONS SET NAME='George' WHERE ID=?; ";
+
+        final Map<String, String> attributes = new HashMap<>();
+        attributes.put("hiveql.args.1.type", String.valueOf(Types.INTEGER));
+        attributes.put("hiveql.args.1.value", "1");
+
+        attributes.put("hiveql.args.2.type", String.valueOf(Types.VARCHAR));
+        attributes.put("hiveql.args.2.value", "Mark");
+
+        attributes.put("hiveql.args.3.type", String.valueOf(Types.INTEGER));
+        attributes.put("hiveql.args.3.value", "84");
+
+        attributes.put("hiveql.args.4.type", String.valueOf(Types.INTEGER));
+        attributes.put("hiveql.args.4.value", "1");
+
+        runner.enqueue(sql.getBytes(), attributes);
+        try {
+            runner.run();
+            fail("Should throw ProcessException");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+        runner.assertAllFlowFilesTransferred(PutHiveQL.REL_RETRY, 0);
+    }
+
     /**
      * Simple implementation only for testing purposes
      */
@@ -607,7 +781,7 @@ public class TestPutHiveQL {
 
         @Override
         public String getConnectionURL() {
-            return service.getConnectionURL();
+            return service != null ? service.getConnectionURL() : null;
         }
     }
 }

nifi-nar-bundles/nifi-hive-bundle/nifi-hive-processors/src/test/java/org/apache/nifi/processors/hive/TestPutHiveStreaming.java

@@ -34,7 +34,7 @@ import org.apache.hive.hcatalog.streaming.StreamingException;
 import org.apache.hive.hcatalog.streaming.TransactionBatch;
 import org.apache.nifi.hadoop.KerberosProperties;
 import org.apache.nifi.hadoop.SecurityUtil;
-import org.apache.nifi.stream.io.ByteArrayOutputStream;
+import org.apache.nifi.processor.exception.ProcessException;
 import org.apache.nifi.util.MockFlowFile;
 import org.apache.nifi.util.TestRunner;
 import org.apache.nifi.util.TestRunners;
@@ -46,6 +46,7 @@ import org.junit.Before;
 import org.junit.Test;
 
 import java.io.ByteArrayInputStream;
+import java.io.ByteArrayOutputStream;
 import java.io.File;
 import java.io.IOException;
 import java.util.Arrays;
@@ -57,11 +58,13 @@ import java.util.Map;
 import java.util.concurrent.ExecutorService;
 
 import static org.apache.nifi.processors.hive.PutHiveStreaming.HIVE_STREAMING_RECORD_COUNT_ATTR;
+import static org.apache.nifi.processors.hive.PutHiveStreaming.REL_SUCCESS;
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertNotNull;
 import static org.junit.Assert.assertNull;
 import static org.junit.Assert.assertTrue;
+import static org.junit.Assert.fail;
 import static org.mockito.Matchers.any;
 import static org.mockito.Matchers.anyLong;
 import static org.mockito.Matchers.anyString;
@@ -213,11 +216,33 @@ public class TestPutHiveStreaming {
     }
 
     @Test
-    public void onTriggerMultipleRecords() throws Exception {
+    public void onTriggerBadInputRollbackOnFailure() throws Exception {
         runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
         runner.setProperty(PutHiveStreaming.DB_NAME, "default");
         runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
-        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "2");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        runner.enqueue("I am not an Avro record".getBytes());
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+    }
+
+
+    @Test
+    public void onTriggerMultipleRecordsSingleTransaction() throws Exception {
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.RECORDS_PER_TXN, "100");
         runner.setValidateExpressionUsage(false);
         Map<String, Object> user1 = new HashMap<String, Object>() {
             {
@@ -237,13 +262,191 @@ public class TestPutHiveStreaming {
                 put("favorite_number", 3);
             }
         };
-        runner.enqueue(createAvroRecord(Arrays.asList(user1, user2, user3)));
+        final List<Map<String, Object>> users = Arrays.asList(user1, user2, user3);
+        runner.enqueue(createAvroRecord(users));
         runner.run();
 
         runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 1);
         MockFlowFile resultFlowFile = runner.getFlowFilesForRelationship(PutHiveStreaming.REL_SUCCESS).get(0);
-        assertNotNull(resultFlowFile);
+        assertOutputAvroRecords(users, resultFlowFile);
-        assertEquals("3", resultFlowFile.getAttribute(PutHiveStreaming.HIVE_STREAMING_RECORD_COUNT_ATTR));
+    }
+
+    @Test
+    public void onTriggerMultipleRecordsMultipleTransaction() throws Exception {
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.RECORDS_PER_TXN, "2");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        Map<String, Object> user2 = new HashMap<String, Object>() {
+            {
+                put("name", "Mary");
+                put("favorite_number", 42);
+            }
+        };
+        Map<String, Object> user3 = new HashMap<String, Object>() {
+            {
+                put("name", "Matt");
+                put("favorite_number", 3);
+            }
+        };
+        final List<Map<String, Object>> users = Arrays.asList(user1, user2, user3);
+        runner.enqueue(createAvroRecord(users));
+        runner.run();
+
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 1);
+        MockFlowFile resultFlowFile = runner.getFlowFilesForRelationship(PutHiveStreaming.REL_SUCCESS).get(0);
+        assertOutputAvroRecords(users, resultFlowFile);
+    }
+
+    @Test
+    public void onTriggerMultipleRecordsFailInMiddle() throws Exception {
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.RECORDS_PER_TXN, "2");
+        runner.setValidateExpressionUsage(false);
+        processor.setGenerateWriteFailure(true, 1);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        Map<String, Object> user2 = new HashMap<String, Object>() {
+            {
+                put("name", "Mary");
+                put("favorite_number", 42);
+            }
+        };
+        Map<String, Object> user3 = new HashMap<String, Object>() {
+            {
+                put("name", "Matt");
+                put("favorite_number", 3);
+            }
+        };
+        Map<String, Object> user4 = new HashMap<String, Object>() {
+            {
+                put("name", "Mike");
+                put("favorite_number", 345);
+            }
+        };
+        final List<Map<String, Object>> users = Arrays.asList(user1, user2, user3, user4);
+        runner.enqueue(createAvroRecord(users));
+        runner.run();
+
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 1);
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 1);
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+
+        MockFlowFile resultFlowFile = runner.getFlowFilesForRelationship(PutHiveStreaming.REL_SUCCESS).get(0);
+        assertOutputAvroRecords(Arrays.asList(user1, user3, user4), resultFlowFile);
+
+        final MockFlowFile failedFlowFile = runner.getFlowFilesForRelationship(PutHiveStreaming.REL_FAILURE).get(0);
+        assertOutputAvroRecords(Arrays.asList(user2), failedFlowFile);
+    }
+
+    @Test
+    public void onTriggerMultipleRecordsFailInMiddleRollbackOnFailure() throws Exception {
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.RECORDS_PER_TXN, "2");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        processor.setGenerateWriteFailure(true, 1);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        Map<String, Object> user2 = new HashMap<String, Object>() {
+            {
+                put("name", "Mary");
+                put("favorite_number", 42);
+            }
+        };
+        Map<String, Object> user3 = new HashMap<String, Object>() {
+            {
+                put("name", "Matt");
+                put("favorite_number", 3);
+            }
+        };
+        runner.enqueue(createAvroRecord(Arrays.asList(user1, user2, user3)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown, because any Hive Transaction is committed yet.");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+    }
+
+    @Test
+    public void onTriggerMultipleRecordsFailInMiddleRollbackOnFailureCommitted() throws Exception {
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.RECORDS_PER_TXN, "2");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        // The first two records are committed, then an issue will happen at the 3rd record.
+        processor.setGenerateWriteFailure(true, 2);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        Map<String, Object> user2 = new HashMap<String, Object>() {
+            {
+                put("name", "Mary");
+                put("favorite_number", 42);
+            }
+        };
+        Map<String, Object> user3 = new HashMap<String, Object>() {
+            {
+                put("name", "Matt");
+                put("favorite_number", 3);
+            }
+        };
+        Map<String, Object> user4 = new HashMap<String, Object>() {
+            {
+                put("name", "Mike");
+                put("favorite_number", 345);
+            }
+        };
+        runner.enqueue(createAvroRecord(Arrays.asList(user1, user2, user3, user4)));
+        // ProcessException should NOT be thrown, because a Hive Transaction is already committed.
+        runner.run();
+
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 1);
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+
+        // Assert transferred FlowFile.
+        assertOutputAvroRecords(Arrays.asList(user1, user2), runner.getFlowFilesForRelationship(REL_SUCCESS).get(0));
+
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+
+    }
+
+    private void assertOutputAvroRecords(List<Map<String, Object>> expectedRecords, MockFlowFile resultFlowFile) throws IOException {
+        assertEquals(String.valueOf(expectedRecords.size()), resultFlowFile.getAttribute(PutHiveStreaming.HIVE_STREAMING_RECORD_COUNT_ATTR));
+
         final DataFileStream<GenericRecord> reader = new DataFileStream<>(
                 new ByteArrayInputStream(resultFlowFile.toByteArray()),
                 new GenericDatumReader<GenericRecord>());
@@ -253,17 +456,20 @@ public class TestPutHiveStreaming {
         // Verify that the schema is preserved
         assertTrue(schema.equals(new Schema.Parser().parse(new File("src/test/resources/user.avsc"))));
 
-        // Verify the records are intact. We can't guarantee order so check the total number and non-null fields
+        GenericRecord record = null;
-        assertTrue(reader.hasNext());
+        for (Map<String, Object> expectedRecord : expectedRecords) {
-        GenericRecord record = reader.next(null);
+            assertTrue(reader.hasNext());
-        assertNotNull(record.get("name"));
+            record = reader.next(record);
-        assertNotNull(record.get("favorite_number"));
+            final String name = record.get("name").toString();
-        assertNull(record.get("favorite_color"));
+            final Integer favorite_number = (Integer) record.get("favorite_number");
-        assertNull(record.get("scale"));
+            assertNotNull(name);
-        assertTrue(reader.hasNext());
+            assertNotNull(favorite_number);
-        record = reader.next(record);
+            assertNull(record.get("favorite_color"));
-        assertTrue(reader.hasNext());
+            assertNull(record.get("scale"));
-        reader.next(record);
+
+            assertEquals(expectedRecord.get("name"), name);
+            assertEquals(expectedRecord.get("favorite_number"), favorite_number);
+        }
         assertFalse(reader.hasNext());
     }
 
@@ -318,6 +524,39 @@ public class TestPutHiveStreaming {
         runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
     }
 
+    @Test
+    public void onTriggerWithPartitionColumnsNotInRecordRollbackOnFailure() throws Exception {
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.PARTITION_COLUMNS, "favorite_food");
+        runner.setProperty(PutHiveStreaming.AUTOCREATE_PARTITIONS, "false");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+                put("favorite_color", "blue");
+            }
+        };
+
+        runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+    }
+
     @Test
     public void onTriggerWithRetireWriters() throws Exception {
         runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
@@ -389,6 +628,36 @@ public class TestPutHiveStreaming {
         runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
     }
 
+    @Test
+    public void onTriggerWithConnectFailureRollbackOnFailure() throws Exception {
+        processor.setGenerateConnectFailure(true);
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+    }
+
     @Test
     public void onTriggerWithInterruptedException() throws Exception {
         processor.setGenerateInterruptedExceptionOnCreateWriter(true);
@@ -409,6 +678,32 @@ public class TestPutHiveStreaming {
         runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 1);
     }
 
+    @Test
+    public void onTriggerWithInterruptedExceptionRollbackOnFailure() throws Exception {
+        processor.setGenerateInterruptedExceptionOnCreateWriter(true);
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+    }
+
     @Test
     public void onTriggerWithWriteFailure() throws Exception {
         processor.setGenerateWriteFailure(true);
@@ -436,6 +731,40 @@ public class TestPutHiveStreaming {
         assertEquals("2", runner.getFlowFilesForRelationship(PutHiveStreaming.REL_FAILURE).get(0).getAttribute(HIVE_STREAMING_RECORD_COUNT_ATTR));
     }
 
+    @Test
+    public void onTriggerWithWriteFailureRollbackOnFailure() throws Exception {
+        processor.setGenerateWriteFailure(true);
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        Map<String, Object> user2 = new HashMap<String, Object>() {
+            {
+                put("name", "Mary");
+                put("favorite_number", 42);
+            }
+        };
+        runner.enqueue(createAvroRecord(Arrays.asList(user1, user2)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+    }
+
     @Test
     public void onTriggerWithSerializationError() throws Exception {
         processor.setGenerateSerializationError(true);
@@ -457,6 +786,35 @@ public class TestPutHiveStreaming {
         runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 1);
     }
 
+    @Test
+    public void onTriggerWithSerializationErrorRollbackOnFailure() throws Exception {
+        processor.setGenerateSerializationError(true);
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
+    }
+
     @Test
     public void onTriggerWithCommitFailure() throws Exception {
         processor.setGenerateCommitFailure(true);
@@ -474,9 +832,39 @@ public class TestPutHiveStreaming {
         runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
         runner.run();
 
-        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 1);
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 1);
+    }
+
+    @Test
+    public void onTriggerWithCommitFailureRollbackOnFailure() throws Exception {
+        processor.setGenerateCommitFailure(true);
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
         runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
         runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
     }
 
     @Test
@@ -496,9 +884,39 @@ public class TestPutHiveStreaming {
         runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
         runner.run();
 
-        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 1);
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
+        runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 1);
+    }
+
+    @Test
+    public void onTriggerWithTransactionFailureRollbackOnFailure() throws Exception {
+        processor.setGenerateTransactionFailure(true);
+        runner.setProperty(PutHiveStreaming.METASTORE_URI, "thrift://localhost:9083");
+        runner.setProperty(PutHiveStreaming.DB_NAME, "default");
+        runner.setProperty(PutHiveStreaming.TABLE_NAME, "users");
+        runner.setProperty(PutHiveStreaming.TXNS_PER_BATCH, "100");
+        runner.setProperty(PutHiveStreaming.ROLLBACK_ON_FAILURE, "true");
+        runner.setValidateExpressionUsage(false);
+        Map<String, Object> user1 = new HashMap<String, Object>() {
+            {
+                put("name", "Joe");
+                put("favorite_number", 146);
+            }
+        };
+        runner.enqueue(createAvroRecord(Collections.singletonList(user1)));
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertTransferCount(PutHiveStreaming.REL_FAILURE, 0);
         runner.assertTransferCount(PutHiveStreaming.REL_SUCCESS, 0);
         runner.assertTransferCount(PutHiveStreaming.REL_RETRY, 0);
+        // Assert incoming FlowFile stays in input queue.
+        assertEquals(1, runner.getQueueSize().getObjectCount());
     }
 
     @Test
@@ -563,10 +981,12 @@ public class TestPutHiveStreaming {
         private boolean generateConnectFailure = false;
         private boolean generateInterruptedExceptionOnCreateWriter = false;
         private boolean generateWriteFailure = false;
+        private Integer generateWriteFailureRecordIndex;
         private boolean generateSerializationError = false;
         private boolean generateCommitFailure = false;
         private boolean generateTransactionFailure = false;
         private boolean generateExceptionOnFlushAndClose = false;
+        private HiveEndPoint hiveEndPoint = mock(HiveEndPoint.class);
 
         @Override
         public KerberosProperties getKerberosProperties() {
@@ -579,7 +999,6 @@ public class TestPutHiveStreaming {
 
         @Override
         public HiveEndPoint makeHiveEndPoint(List<String> partitionValues, HiveOptions hiveOptions) {
-            HiveEndPoint hiveEndPoint = mock(HiveEndPoint.class);
             return hiveEndPoint;
         }
 
@@ -593,7 +1012,7 @@ public class TestPutHiveStreaming {
                 throw new InterruptedException();
             }
             MockHiveWriter hiveWriter = new MockHiveWriter(endPoint, options.getTxnsPerBatch(), options.getAutoCreatePartitions(), options.getCallTimeOut(), callTimeoutPool, ugi, hiveConfig);
-            hiveWriter.setGenerateWriteFailure(generateWriteFailure);
+            hiveWriter.setGenerateWriteFailure(generateWriteFailure, generateWriteFailureRecordIndex);
             hiveWriter.setGenerateSerializationError(generateSerializationError);
             hiveWriter.setGenerateCommitFailure(generateCommitFailure);
             hiveWriter.setGenerateTransactionFailure(generateTransactionFailure);
@@ -613,6 +1032,11 @@ public class TestPutHiveStreaming {
             this.generateWriteFailure = generateWriteFailure;
         }
 
+        public void setGenerateWriteFailure(boolean generateWriteFailure, int generateWriteFailureRecordIndex) {
+            this.generateWriteFailure = generateWriteFailure;
+            this.generateWriteFailureRecordIndex = generateWriteFailureRecordIndex;
+        }
+
         public void setGenerateSerializationError(boolean generateSerializationError) {
             this.generateSerializationError = generateSerializationError;
         }
@@ -634,10 +1058,13 @@ public class TestPutHiveStreaming {
     private class MockHiveWriter extends HiveWriter {
 
         private boolean generateWriteFailure = false;
+        private Integer generateWriteFailureRecordIndex;
         private boolean generateSerializationError = false;
         private boolean generateCommitFailure = false;
         private boolean generateTransactionFailure = false;
         private boolean generateExceptionOnFlushAndClose = false;
+        private int writeAttemptCount = 0;
+        private int totalRecords = 0;
 
         private HiveEndPoint endPoint;
 
@@ -651,16 +1078,23 @@ public class TestPutHiveStreaming {
 
         @Override
         public synchronized void write(byte[] record) throws WriteFailure, SerializationError, InterruptedException {
-            if (generateWriteFailure) {
+            try {
-                throw new HiveWriter.WriteFailure(endPoint, 1L, new Exception());
+                if (generateWriteFailure
-            }
+                        && (generateWriteFailureRecordIndex == null || writeAttemptCount == generateWriteFailureRecordIndex)) {
-            if (generateSerializationError) {
+                    throw new WriteFailure(endPoint, 1L, new Exception());
-                throw new SerializationError("Test Serialization Error", new Exception());
+                }
+                if (generateSerializationError) {
+                    throw new SerializationError("Test Serialization Error", new Exception());
+                }
+                totalRecords++;
+            } finally {
+                writeAttemptCount++;
             }
         }
 
-        public void setGenerateWriteFailure(boolean generateWriteFailure) {
+        public void setGenerateWriteFailure(boolean generateWriteFailure, Integer generateWriteFailureRecordIndex) {
             this.generateWriteFailure = generateWriteFailure;
+            this.generateWriteFailureRecordIndex = generateWriteFailureRecordIndex;
         }
 
         public void setGenerateSerializationError(boolean generateSerializationError) {
@@ -754,6 +1188,11 @@ public class TestPutHiveStreaming {
         protected void nextTxn(boolean rollToNext) throws StreamingException, InterruptedException, TxnBatchFailure {
             // Empty
         }
+
+        @Override
+        public int getTotalRecords() {
+            return totalRecords;
+        }
     }
 
 }

nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/PutSQL.java

@@ -26,18 +26,28 @@ import org.apache.nifi.annotation.behavior.WritesAttributes;
 import org.apache.nifi.annotation.documentation.CapabilityDescription;
 import org.apache.nifi.annotation.documentation.SeeAlso;
 import org.apache.nifi.annotation.documentation.Tags;
+import org.apache.nifi.annotation.lifecycle.OnScheduled;
 import org.apache.nifi.components.PropertyDescriptor;
 import org.apache.nifi.dbcp.DBCPService;
 import org.apache.nifi.flowfile.FlowFile;
 import org.apache.nifi.flowfile.attributes.FragmentAttributes;
-import org.apache.nifi.processor.AbstractProcessor;
+import org.apache.nifi.processor.AbstractSessionFactoryProcessor;
 import org.apache.nifi.processor.FlowFileFilter;
 import org.apache.nifi.processor.ProcessContext;
 import org.apache.nifi.processor.ProcessSession;
+import org.apache.nifi.processor.ProcessSessionFactory;
 import org.apache.nifi.processor.Relationship;
 import org.apache.nifi.processor.exception.ProcessException;
 import org.apache.nifi.processor.io.InputStreamCallback;
 import org.apache.nifi.processor.util.StandardValidators;
+import org.apache.nifi.processor.util.pattern.ErrorTypes;
+import org.apache.nifi.processor.util.pattern.ExceptionHandler;
+import org.apache.nifi.processor.util.pattern.PartialFunctions;
+import org.apache.nifi.processor.util.pattern.PartialFunctions.FetchFlowFiles;
+import org.apache.nifi.processor.util.pattern.PartialFunctions.FlowFileGroup;
+import org.apache.nifi.processor.util.pattern.PutGroup;
+import org.apache.nifi.processor.util.pattern.RollbackOnFailure;
+import org.apache.nifi.processor.util.pattern.RoutingResult;
 import org.apache.nifi.stream.io.StreamUtils;
 
 import javax.xml.bind.DatatypeConverter;
@@ -52,6 +62,7 @@ import java.sql.Connection;
 import java.sql.Date;
 import java.sql.PreparedStatement;
 import java.sql.ResultSet;
+import java.sql.SQLDataException;
 import java.sql.SQLException;
 import java.sql.SQLNonTransientException;
 import java.sql.Statement;
@@ -65,19 +76,19 @@ import java.time.format.DateTimeFormatter;
 import java.time.temporal.TemporalAccessor;
 import java.util.ArrayList;
 import java.util.BitSet;
-import java.util.Collections;
 import java.util.Comparator;
 import java.util.HashMap;
 import java.util.HashSet;
-import java.util.LinkedHashSet;
 import java.util.List;
-import java.util.ListIterator;
 import java.util.Map;
 import java.util.Set;
 import java.util.concurrent.TimeUnit;
+import java.util.function.BiFunction;
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
+import static org.apache.nifi.processor.util.pattern.ExceptionHandler.createOnError;
+
 @SupportsBatching
 @SeeAlso(ConvertJSONToSQL.class)
 @InputRequirement(Requirement.INPUT_REQUIRED)
@@ -109,7 +120,7 @@ import java.util.regex.Pattern;
         @WritesAttribute(attribute = "sql.generated.key", description = "If the database generated a key for an INSERT statement and the Obtain Generated Keys property is set to true, "
                 + "this attribute will be added to indicate the generated key, if possible. This feature is not supported by all database vendors.")
 })
-public class PutSQL extends AbstractProcessor {
+public class PutSQL extends AbstractSessionFactoryProcessor {
 
     static final PropertyDescriptor CONNECTION_POOL = new PropertyDescriptor.Builder()
             .name("JDBC Connection Pool")
@@ -180,6 +191,7 @@ public class PutSQL extends AbstractProcessor {
         properties.add(TRANSACTION_TIMEOUT);
         properties.add(BATCH_SIZE);
         properties.add(OBTAIN_GENERATED_KEYS);
+        properties.add(RollbackOnFailure.ROLLBACK_ON_FAILURE);
         return properties;
     }
 
@@ -192,199 +204,176 @@ public class PutSQL extends AbstractProcessor {
         return rels;
     }
 
+    private static class FunctionContext extends RollbackOnFailure {
+        private boolean obtainKeys = false;
+        private boolean fragmentedTransaction = false;
+        private boolean originalAutoCommit = false;
+        private final long startNanos = System.nanoTime();
+
+        private FunctionContext(boolean rollbackOnFailure) {
+            super(rollbackOnFailure, true);
+        }
+
+        private boolean isSupportBatching() {
+            return !obtainKeys && !fragmentedTransaction;
+        }
+    }
+
+    private PutGroup<FunctionContext, Connection, StatementFlowFileEnclosure> process;
+    private BiFunction<FunctionContext, ErrorTypes, ErrorTypes.Result> adjustError;
+    private ExceptionHandler<FunctionContext> exceptionHandler;
 
 
-    @Override
+    private final FetchFlowFiles<FunctionContext> fetchFlowFiles = (c, s, fc, r) -> {
-    public void onTrigger(final ProcessContext context, final ProcessSession session) throws ProcessException {
+        final FlowFilePoll poll = pollFlowFiles(c, s, fc, r);
-        final FlowFilePoll poll = pollFlowFiles(context, session);
         if (poll == null) {
-            return;
+            return null;
         }
+        fc.fragmentedTransaction = poll.isFragmentedTransaction();
+        return poll.getFlowFiles();
+    };
 
-        final List<FlowFile> flowFiles = poll.getFlowFiles();
+    private final PartialFunctions.InitConnection<FunctionContext, Connection> initConnection = (c, s, fc) -> {
-        if (flowFiles == null) {
+        final Connection connection = c.getProperty(CONNECTION_POOL).asControllerService(DBCPService.class).getConnection();
-            return;
+        try {
+            fc.originalAutoCommit = connection.getAutoCommit();
+            connection.setAutoCommit(false);
+        } catch (SQLException e) {
+            throw new ProcessException("Failed to disable auto commit due to " + e, e);
         }
+        return connection;
+    };
 
-        final long startNanos = System.nanoTime();
-        final boolean obtainKeys = context.getProperty(OBTAIN_GENERATED_KEYS).asBoolean();
-        final Map<String, StatementFlowFileEnclosure> statementMap = new HashMap<>(); // Map SQL to a PreparedStatement and FlowFiles
-        final List<FlowFile> sentFlowFiles = new ArrayList<>(); // flowfiles that have been sent
-        final List<FlowFile> processedFlowFiles = new ArrayList<>(); // all flowfiles that we have processed
-        final Set<StatementFlowFileEnclosure> enclosuresToExecute = new LinkedHashSet<>(); // the enclosures that we've processed
 
-        // Because we can have a transaction that is necessary across multiple FlowFiles, things get complicated when
+    @FunctionalInterface
-        // some FlowFiles have been transferred to a relationship and then there is a failure. As a result, we will just
+    private interface GroupingFunction {
-        // map all FlowFiles to their destination relationship and do the session.transfer at the end. This way, if there
+        void apply(final ProcessContext context, final ProcessSession session, final FunctionContext fc,
-        // is a failure, we can route all FlowFiles to failure if we need to.
+                   final Connection conn, final List<FlowFile> flowFiles,
-        final Map<FlowFile, Relationship> destinationRelationships = new HashMap<>();
+                   final List<StatementFlowFileEnclosure> groups,
+                   final Map<String, StatementFlowFileEnclosure> sqlToEnclosure,
+                   final RoutingResult result);
+    }
 
-        final DBCPService dbcpService = context.getProperty(CONNECTION_POOL).asControllerService(DBCPService.class);
+    private GroupingFunction groupFragmentedTransaction = (context, session, fc, conn, flowFiles, groups, sqlToEnclosure, result) -> {
-        try (final Connection conn = dbcpService.getConnection()) {
+        final FragmentedEnclosure fragmentedEnclosure = new FragmentedEnclosure();
-            final boolean originalAutoCommit = conn.getAutoCommit();
+        groups.add(fragmentedEnclosure);
-            try {
-                conn.setAutoCommit(false);
 
-                for (final FlowFile flowFile : flowFiles) {
+        for (final FlowFile flowFile : flowFiles) {
-                    processedFlowFiles.add(flowFile);
+            final String sql = getSQL(session, flowFile);
-                    final String sql = getSQL(session, flowFile);
 
-                    // Get the appropriate PreparedStatement to use.
+            final StatementFlowFileEnclosure enclosure = sqlToEnclosure
-                    final StatementFlowFileEnclosure enclosure;
+                    .computeIfAbsent(sql, k -> new StatementFlowFileEnclosure(sql));
-                    try {
-                        enclosure = getEnclosure(sql, conn, statementMap, obtainKeys, poll.isFragmentedTransaction());
-                    } catch (final SQLNonTransientException e) {
-                        getLogger().error("Failed to update database for {} due to {}; routing to failure", new Object[] {flowFile, e});
-                        destinationRelationships.put(flowFile, REL_FAILURE);
-                        continue;
-                    }
 
-                    final PreparedStatement stmt = enclosure.getStatement();
+            fragmentedEnclosure.addFlowFile(flowFile, enclosure);
+        }
+    };
 
-                    // set the appropriate parameters on the statement.
+    private final GroupingFunction groupFlowFilesBySQLBatch = (context, session, fc, conn, flowFiles, groups, sqlToEnclosure, result) -> {
-                    try {
+        for (final FlowFile flowFile : flowFiles) {
-                        setParameters(stmt, flowFile.getAttributes());
+            final String sql = getSQL(session, flowFile);
-                    } catch (final SQLException | ProcessException pe) {
-                        getLogger().error("Cannot update database for {} due to {}; routing to failure", new Object[] {flowFile, pe.toString()}, pe);
-                        destinationRelationships.put(flowFile, REL_FAILURE);
-                        continue;
-                    }
 
-                    // If we need to obtain keys, we cannot do so in a a Batch Update. So we have to execute the statement and close it.
+            // Get or create the appropriate PreparedStatement to use.
-                    if (obtainKeys) {
+            final StatementFlowFileEnclosure enclosure = sqlToEnclosure
-                        try {
+                    .computeIfAbsent(sql, k -> {
-                            // Execute the actual update.
+                        final StatementFlowFileEnclosure newEnclosure = new StatementFlowFileEnclosure(sql);
-                            stmt.executeUpdate();
+                        groups.add(newEnclosure);
+                        return newEnclosure;
+                    });
 
-                            // attempt to determine the key that was generated, if any. This is not supported by all
+            if(!exceptionHandler.execute(fc, flowFile, input -> {
-                            // database vendors, so if we cannot determine the generated key (or if the statement is not an INSERT),
+                final PreparedStatement stmt = enclosure.getCachedStatement(conn);
-                            // we will just move on without setting the attribute.
+                setParameters(stmt, flowFile.getAttributes());
-                            FlowFile sentFlowFile = flowFile;
+                stmt.addBatch();
-                            final String generatedKey = determineGeneratedKey(stmt);
+            }, onFlowFileError(context, session, result))) {
-                            if (generatedKey != null) {
+                continue;
-                                sentFlowFile = session.putAttribute(sentFlowFile, "sql.generated.key", generatedKey);
-                            }
-
-                            stmt.close();
-                            sentFlowFiles.add(sentFlowFile);
-                        } catch (final SQLNonTransientException e) {
-                            getLogger().error("Failed to update database for {} due to {}; routing to failure", new Object[] {flowFile, e});
-                            destinationRelationships.put(flowFile, REL_FAILURE);
-                            continue;
-                        } catch (final SQLException e) {
-                            getLogger().error("Failed to update database for {} due to {}; it is possible that retrying the operation will succeed, so routing to retry", new Object[] {flowFile, e});
-                            destinationRelationships.put(flowFile, REL_RETRY);
-                            continue;
-                        }
-                    } else {
-                        // We don't need to obtain keys. Just add the statement to the batch.
-                        stmt.addBatch();
-                        enclosure.addFlowFile(flowFile);
-                        enclosuresToExecute.add(enclosure);
-                    }
-                }
-
-                // If we are not trying to obtain the generated keys, we will have
-                // PreparedStatement's that have batches added to them. We need to execute each batch and close
-                // the PreparedStatement.
-                for (final StatementFlowFileEnclosure enclosure : enclosuresToExecute) {
-                    try {
-                        final PreparedStatement stmt = enclosure.getStatement();
-                        stmt.executeBatch();
-                        sentFlowFiles.addAll(enclosure.getFlowFiles());
-                    } catch (final BatchUpdateException e) {
-                        // If we get a BatchUpdateException, then we want to determine which FlowFile caused the failure,
-                        // and route that FlowFile to failure while routing those that finished processing to success and those
-                        // that have not yet been executed to retry. If the FlowFile was
-                        // part of a fragmented transaction, then we must roll back all updates for this connection, because
-                        // other statements may have been successful and been part of this transaction.
-                        final int[] updateCounts = e.getUpdateCounts();
-                        final int offendingFlowFileIndex = updateCounts.length;
-                        final List<FlowFile> batchFlowFiles = enclosure.getFlowFiles();
-
-                        if (poll.isFragmentedTransaction()) {
-                            // There are potentially multiple statements for this one transaction. As a result,
-                            // we need to roll back the entire transaction and route all of the FlowFiles to failure.
-                            conn.rollback();
-                            final FlowFile offendingFlowFile = batchFlowFiles.get(offendingFlowFileIndex);
-                            getLogger().error("Failed to update database due to a failed batch update. A total of {} FlowFiles are required for this transaction, so routing all to failure. "
-                                    + "Offending FlowFile was {}, which caused the following error: {}", new Object[] {flowFiles.size(), offendingFlowFile, e});
-                            session.transfer(flowFiles, REL_FAILURE);
-                            return;
-                        }
-
-                        // In the presence of a BatchUpdateException, the driver has the option of either stopping when an error
-                        // occurs, or continuing. If it continues, then it must account for all statements in the batch and for
-                        // those that fail return a Statement.EXECUTE_FAILED for the number of rows updated.
-                        // So we will iterate over all of the update counts returned. If any is equal to Statement.EXECUTE_FAILED,
-                        // we will route the corresponding FlowFile to failure. Otherwise, the FlowFile will go to success
-                        // unless it has not yet been processed (its index in the List > updateCounts.length).
-                        int failureCount = 0;
-                        int successCount = 0;
-                        int retryCount = 0;
-                        for (int i = 0; i < updateCounts.length; i++) {
-                            final int updateCount = updateCounts[i];
-                            final FlowFile flowFile = batchFlowFiles.get(i);
-                            if (updateCount == Statement.EXECUTE_FAILED) {
-                                destinationRelationships.put(flowFile, REL_FAILURE);
-                                failureCount++;
-                            } else {
-                                destinationRelationships.put(flowFile, REL_SUCCESS);
-                                successCount++;
-                            }
-                        }
-
-                        if (failureCount == 0) {
-                            // if no failures found, the driver decided not to execute the statements after the
-                            // failure, so route the last one to failure.
-                            final FlowFile failedFlowFile = batchFlowFiles.get(updateCounts.length);
-                            destinationRelationships.put(failedFlowFile, REL_FAILURE);
-                            failureCount++;
-                        }
-
-                        if (updateCounts.length < batchFlowFiles.size()) {
-                            final List<FlowFile> unexecuted = batchFlowFiles.subList(updateCounts.length + 1, batchFlowFiles.size());
-                            for (final FlowFile flowFile : unexecuted) {
-                                destinationRelationships.put(flowFile, REL_RETRY);
-                                retryCount++;
-                            }
-                        }
-
-                        getLogger().error("Failed to update database due to a failed batch update. There were a total of {} FlowFiles that failed, {} that succeeded, "
-                                + "and {} that were not execute and will be routed to retry; ", new Object[] {failureCount, successCount, retryCount});
-                    } catch (final SQLNonTransientException e) {
-                        getLogger().error("Failed to update database for {} due to {}; routing to failure", new Object[] {enclosure.getFlowFiles(), e});
-
-                        for (final FlowFile flowFile : enclosure.getFlowFiles()) {
-                            destinationRelationships.put(flowFile, REL_FAILURE);
-                        }
-                        continue;
-                    } catch (final SQLException e) {
-                        getLogger().error("Failed to update database for {} due to {}; it is possible that retrying the operation will succeed, so routing to retry",
-                                new Object[] {enclosure.getFlowFiles(), e});
-
-                        for (final FlowFile flowFile : enclosure.getFlowFiles()) {
-                            destinationRelationships.put(flowFile, REL_RETRY);
-                        }
-                        continue;
-                    } finally {
-                        enclosure.getStatement().close();
-                    }
-                }
-            } finally {
-                try {
-                    conn.commit();
-                } finally {
-                    // make sure that we try to set the auto commit back to whatever it was.
-                    if (originalAutoCommit) {
-                        try {
-                            conn.setAutoCommit(originalAutoCommit);
-                        } catch (final SQLException se) {
-                        }
-                    }
-                }
             }
 
+            enclosure.addFlowFile(flowFile);
+        }
+    };
+
+    private GroupingFunction groupFlowFilesBySQL = (context, session, fc, conn, flowFiles, groups, sqlToEnclosure, result) -> {
+        for (final FlowFile flowFile : flowFiles) {
+            final String sql = getSQL(session, flowFile);
+
+            // Get or create the appropriate PreparedStatement to use.
+            final StatementFlowFileEnclosure enclosure = sqlToEnclosure
+                    .computeIfAbsent(sql, k -> {
+                        final StatementFlowFileEnclosure newEnclosure = new StatementFlowFileEnclosure(sql);
+                        groups.add(newEnclosure);
+                        return newEnclosure;
+                    });
+
+            enclosure.addFlowFile(flowFile);
+        }
+    };
+
+    final PutGroup.GroupFlowFiles<FunctionContext, Connection, StatementFlowFileEnclosure> groupFlowFiles = (context, session, fc, conn, flowFiles, result) -> {
+        final Map<String, StatementFlowFileEnclosure> sqlToEnclosure = new HashMap<>();
+        final List<StatementFlowFileEnclosure> groups = new ArrayList<>();
+
+        // There are three patterns:
+        // 1. Support batching: An enclosure has multiple FlowFiles being executed in a batch operation
+        // 2. Obtain keys: An enclosure has multiple FlowFiles, and each FlowFile is executed separately
+        // 3. Fragmented transaction: One FlowFile per Enclosure?
+        if (fc.obtainKeys) {
+            groupFlowFilesBySQL.apply(context, session, fc, conn, flowFiles, groups, sqlToEnclosure, result);
+        } else if (fc.fragmentedTransaction) {
+            groupFragmentedTransaction.apply(context, session, fc, conn, flowFiles, groups, sqlToEnclosure, result);
+        } else {
+            groupFlowFilesBySQLBatch.apply(context, session, fc, conn, flowFiles, groups, sqlToEnclosure, result);
+        }
+
+        return groups;
+    };
+
+    final PutGroup.PutFlowFiles<FunctionContext, Connection, StatementFlowFileEnclosure> putFlowFiles = (context, session, fc, conn, enclosure, result) -> {
+
+        if (fc.isSupportBatching()) {
+
+            // We have PreparedStatement that have batches added to them.
+            // We need to execute each batch and close the PreparedStatement.
+            exceptionHandler.execute(fc, enclosure, input -> {
+                try (final PreparedStatement stmt = enclosure.getCachedStatement(conn)) {
+                    stmt.executeBatch();
+                    result.routeTo(enclosure.getFlowFiles(), REL_SUCCESS);
+                }
+            }, onBatchUpdateError(context, session, result));
+
+        } else {
+            for (final FlowFile flowFile : enclosure.getFlowFiles()) {
+
+                final StatementFlowFileEnclosure targetEnclosure
+                        = enclosure instanceof FragmentedEnclosure
+                        ? ((FragmentedEnclosure) enclosure).getTargetEnclosure(flowFile)
+                        : enclosure;
+
+                // Execute update one by one.
+                exceptionHandler.execute(fc, flowFile, input -> {
+                    try (final PreparedStatement stmt = targetEnclosure.getNewStatement(conn, fc.obtainKeys)) {
+
+                        // set the appropriate parameters on the statement.
+                        setParameters(stmt, flowFile.getAttributes());
+
+                        stmt.executeUpdate();
+
+                        // attempt to determine the key that was generated, if any. This is not supported by all
+                        // database vendors, so if we cannot determine the generated key (or if the statement is not an INSERT),
+                        // we will just move on without setting the attribute.
+                        FlowFile sentFlowFile = flowFile;
+                        final String generatedKey = determineGeneratedKey(stmt);
+                        if (generatedKey != null) {
+                            sentFlowFile = session.putAttribute(sentFlowFile, "sql.generated.key", generatedKey);
+                        }
+
+                        result.routeTo(sentFlowFile, REL_SUCCESS);
+
+                    }
+                }, onFlowFileError(context, session, result));
+            }
+        }
+
+        if (result.contains(REL_SUCCESS)) {
             // Determine the database URL
             String url = "jdbc://unknown-host";
             try {
@@ -393,46 +382,167 @@ public class PutSQL extends AbstractProcessor {
             }
 
             // Emit a Provenance SEND event
-            final long transmissionMillis = TimeUnit.NANOSECONDS.toMillis(System.nanoTime() - startNanos);
+            final long transmissionMillis = TimeUnit.NANOSECONDS.toMillis(System.nanoTime() - fc.startNanos);
-            for (final FlowFile flowFile : sentFlowFiles) {
+            for (final FlowFile flowFile : result.getRoutedFlowFiles().get(REL_SUCCESS)) {
                 session.getProvenanceReporter().send(flowFile, url, transmissionMillis, true);
             }
-
-            for (final FlowFile flowFile : sentFlowFiles) {
-                destinationRelationships.put(flowFile, REL_SUCCESS);
-            }
-        } catch (final SQLException e) {
-            // Failed FlowFiles are all of them that we have processed minus those that were successfully sent
-            final List<FlowFile> failedFlowFiles = processedFlowFiles;
-            failedFlowFiles.removeAll(sentFlowFiles);
-
-            // All FlowFiles yet to be processed is all FlowFiles minus those processed
-            final List<FlowFile> retry = flowFiles;
-            retry.removeAll(processedFlowFiles);
-
-            final Relationship rel;
-            if (e instanceof SQLNonTransientException) {
-                getLogger().error("Failed to update database for {} due to {}; routing to failure", new Object[] {failedFlowFiles, e});
-                rel = REL_FAILURE;
-            } else {
-                getLogger().error("Failed to update database for {} due to {}; it is possible that retrying the operation will succeed, so routing to retry", new Object[] {failedFlowFiles, e});
-                rel = REL_RETRY;
-            }
-
-            for (final FlowFile flowFile : failedFlowFiles) {
-                destinationRelationships.put(flowFile, rel);
-            }
-
-            for (final FlowFile flowFile : retry) {
-                destinationRelationships.put(flowFile, Relationship.SELF);
-            }
         }
+    };
 
-        for (final Map.Entry<FlowFile, Relationship> entry : destinationRelationships.entrySet()) {
+    private ExceptionHandler.OnError<FunctionContext, FlowFile> onFlowFileError(final ProcessContext context, final ProcessSession session, final RoutingResult result) {
-            session.transfer(entry.getKey(), entry.getValue());
+        ExceptionHandler.OnError<FunctionContext, FlowFile> onFlowFileError = createOnError(context, session, result, REL_FAILURE, REL_RETRY);
-        }
+        onFlowFileError = onFlowFileError.andThen((c, i, r, e) -> {
+            switch (r.destination()) {
+                case Failure:
+                    getLogger().error("Failed to update database for {} due to {}; routing to failure", new Object[] {i, e}, e);
+                    break;
+                case Retry:
+                    getLogger().error("Failed to update database for {} due to {}; it is possible that retrying the operation will succeed, so routing to retry",
+                            new Object[] {i, e}, e);
+                    break;
+            }
+        });
+        return RollbackOnFailure.createOnError(onFlowFileError);
     }
 
+    private ExceptionHandler.OnError<FunctionContext, StatementFlowFileEnclosure> onBatchUpdateError(
+            final ProcessContext context, final ProcessSession session, final RoutingResult result) {
+        return RollbackOnFailure.createOnError((c, enclosure, r, e) -> {
+
+            // If rollbackOnFailure is enabled, the error will be thrown as ProcessException instead.
+            if (e instanceof BatchUpdateException && !c.isRollbackOnFailure()) {
+
+                // If we get a BatchUpdateException, then we want to determine which FlowFile caused the failure,
+                // and route that FlowFile to failure while routing those that finished processing to success and those
+                // that have not yet been executed to retry.
+                // Currently fragmented transaction does not use batch update.
+                final int[] updateCounts = ((BatchUpdateException) e).getUpdateCounts();
+                final List<FlowFile> batchFlowFiles = enclosure.getFlowFiles();
+
+                // In the presence of a BatchUpdateException, the driver has the option of either stopping when an error
+                // occurs, or continuing. If it continues, then it must account for all statements in the batch and for
+                // those that fail return a Statement.EXECUTE_FAILED for the number of rows updated.
+                // So we will iterate over all of the update counts returned. If any is equal to Statement.EXECUTE_FAILED,
+                // we will route the corresponding FlowFile to failure. Otherwise, the FlowFile will go to success
+                // unless it has not yet been processed (its index in the List > updateCounts.length).
+                int failureCount = 0;
+                int successCount = 0;
+                int retryCount = 0;
+                for (int i = 0; i < updateCounts.length; i++) {
+                    final int updateCount = updateCounts[i];
+                    final FlowFile flowFile = batchFlowFiles.get(i);
+                    if (updateCount == Statement.EXECUTE_FAILED) {
+                        result.routeTo(flowFile, REL_FAILURE);
+                        failureCount++;
+                    } else {
+                        result.routeTo(flowFile, REL_SUCCESS);
+                        successCount++;
+                    }
+                }
+
+                if (failureCount == 0) {
+                    // if no failures found, the driver decided not to execute the statements after the
+                    // failure, so route the last one to failure.
+                    final FlowFile failedFlowFile = batchFlowFiles.get(updateCounts.length);
+                    result.routeTo(failedFlowFile, REL_FAILURE);
+                    failureCount++;
+                }
+
+                if (updateCounts.length < batchFlowFiles.size()) {
+                    final List<FlowFile> unexecuted = batchFlowFiles.subList(updateCounts.length + 1, batchFlowFiles.size());
+                    for (final FlowFile flowFile : unexecuted) {
+                        result.routeTo(flowFile, REL_RETRY);
+                        retryCount++;
+                    }
+                }
+
+                getLogger().error("Failed to update database due to a failed batch update. There were a total of {} FlowFiles that failed, {} that succeeded, "
+                        + "and {} that were not execute and will be routed to retry; ", new Object[]{failureCount, successCount, retryCount});
+
+                return;
+
+            }
+
+            // Apply default error handling and logging for other Exceptions.
+            ExceptionHandler.OnError<RollbackOnFailure, FlowFileGroup> onGroupError
+                    = ExceptionHandler.createOnGroupError(context, session, result, REL_FAILURE, REL_RETRY);
+            onGroupError = onGroupError.andThen((cl, il, rl, el) -> {
+                switch (r.destination()) {
+                    case Failure:
+                        getLogger().error("Failed to update database for {} due to {}; routing to failure", new Object[] {il.getFlowFiles(), e}, e);
+                        break;
+                    case Retry:
+                        getLogger().error("Failed to update database for {} due to {}; it is possible that retrying the operation will succeed, so routing to retry",
+                                new Object[] {il.getFlowFiles(), e}, e);
+                        break;
+                }
+            });
+            onGroupError.apply(c, enclosure, r, e);
+        });
+    }
+
+    @OnScheduled
+    public void constructProcess() {
+        process = new PutGroup<>();
+
+        process.setLogger(getLogger());
+        process.fetchFlowFiles(fetchFlowFiles);
+        process.initConnection(initConnection);
+        process.groupFetchedFlowFiles(groupFlowFiles);
+        process.putFlowFiles(putFlowFiles);
+        process.adjustRoute(RollbackOnFailure.createAdjustRoute(REL_FAILURE, REL_RETRY));
+
+        process.onCompleted((c, s, fc, conn) -> {
+            try {
+                conn.commit();
+            } catch (SQLException e) {
+                // Throw ProcessException to rollback process session.
+                throw new ProcessException("Failed to commit database connection due to " + e, e);
+            }
+        });
+
+        process.onFailed((c, s, fc, conn, e) -> {
+            try {
+                conn.rollback();
+            } catch (SQLException re) {
+                // Just log the fact that rollback failed.
+                // ProcessSession will be rollback by the thrown Exception so don't have to do anything here.
+                getLogger().warn("Failed to rollback database connection due to %s", new Object[]{re}, re);
+            }
+        });
+
+        process.cleanup((c, s, fc, conn) -> {
+            // make sure that we try to set the auto commit back to whatever it was.
+            if (fc.originalAutoCommit) {
+                try {
+                    conn.setAutoCommit(true);
+                } catch (final SQLException se) {
+                    getLogger().warn("Failed to reset autocommit due to {}", new Object[]{se});
+                }
+            }
+        });
+
+        exceptionHandler = new ExceptionHandler<>();
+        exceptionHandler.mapException(e -> {
+            if (e instanceof SQLNonTransientException) {
+                return ErrorTypes.InvalidInput;
+            } else if (e instanceof SQLException) {
+                return ErrorTypes.TemporalFailure;
+            } else {
+                return ErrorTypes.UnknownFailure;
+            }
+        });
+        adjustError = RollbackOnFailure.createAdjustError(getLogger());
+        exceptionHandler.adjustError(adjustError);
+    }
+
+    @Override
+    public void onTrigger(ProcessContext context, ProcessSessionFactory sessionFactory) throws ProcessException {
+        final Boolean rollbackOnFailure = context.getProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE).asBoolean();
+        final FunctionContext functionContext = new FunctionContext(rollbackOnFailure);
+        functionContext.obtainKeys = context.getProperty(OBTAIN_GENERATED_KEYS).asBoolean();
+        RollbackOnFailure.onTrigger(context, sessionFactory, functionContext, getLogger(), session -> process.onTrigger(context, session, functionContext));
+    }
 
     /**
      * Pulls a batch of FlowFiles from the incoming queues. If no FlowFiles are available, returns <code>null</code>.
@@ -448,7 +558,8 @@ public class PutSQL extends AbstractProcessor {
      * @param session the process session for pulling flowfiles
      * @return a FlowFilePoll containing a List of FlowFiles to process, or <code>null</code> if there are no FlowFiles to process
      */
-    private FlowFilePoll pollFlowFiles(final ProcessContext context, final ProcessSession session) {
+    private FlowFilePoll pollFlowFiles(final ProcessContext context, final ProcessSession session,
+                                       final FunctionContext functionContext, final RoutingResult result) {
         // Determine which FlowFile Filter to use in order to obtain FlowFiles.
         final boolean useTransactions = context.getProperty(SUPPORT_TRANSACTIONS).asBoolean();
         boolean fragmentedTransaction = false;
@@ -469,31 +580,23 @@ public class PutSQL extends AbstractProcessor {
 
         // If we are supporting fragmented transactions, verify that all FlowFiles are correct
         if (fragmentedTransaction) {
-            final Relationship relationship = determineRelationship(flowFiles, context.getProperty(TRANSACTION_TIMEOUT).asTimePeriod(TimeUnit.MILLISECONDS));
+            try {
-            if (relationship != null) {
+                if (!isFragmentedTransactionReady(flowFiles, context.getProperty(TRANSACTION_TIMEOUT).asTimePeriod(TimeUnit.MILLISECONDS))) {
-                // if transferring back to self, penalize the FlowFiles.
+                    // Not ready, penalize FlowFiles and put it back to self.
-                if (relationship == Relationship.SELF) {
+                    flowFiles.forEach(f -> result.routeTo(session.penalize(f), Relationship.SELF));
-                    // penalize all of the FlowFiles that we are going to route to SELF.
+                    return null;
-                    final ListIterator<FlowFile> itr = flowFiles.listIterator();
-                    while (itr.hasNext()) {
-                        final FlowFile flowFile = itr.next();
-                        final FlowFile penalized = session.penalize(flowFile);
-                        itr.remove();
-                        itr.add(penalized);
-                    }
                 }
 
-                session.transfer(flowFiles, relationship);
+            } catch (IllegalArgumentException e) {
+                // Map relationship based on context, and then let default handler to handle.
+                final ErrorTypes.Result adjustedRoute = adjustError.apply(functionContext, ErrorTypes.InvalidInput);
+                ExceptionHandler.createOnGroupError(context, session, result, REL_FAILURE, REL_RETRY)
+                        .apply(functionContext, () -> flowFiles, adjustedRoute, e);
                 return null;
             }
 
             // sort by fragment index.
-            Collections.sort(flowFiles, new Comparator<FlowFile>() {
+            flowFiles.sort(Comparator.comparing(o -> Integer.parseInt(o.getAttribute(FRAGMENT_INDEX_ATTR))));
-                @Override
-                public int compare(final FlowFile o1, final FlowFile o2) {
-                    return Integer.compare(Integer.parseInt(o1.getAttribute(FRAGMENT_INDEX_ATTR)), Integer.parseInt(o2.getAttribute(FRAGMENT_INDEX_ATTR)));
-                }
-            });
         }
 
         return new FlowFilePoll(flowFiles, fragmentedTransaction);
@@ -521,63 +624,6 @@ public class PutSQL extends AbstractProcessor {
         return null;
     }
 
-
-    /**
-     * Returns the StatementFlowFileEnclosure that should be used for executing the given SQL statement
-     *
-     * @param sql the SQL to execute
-     * @param conn the connection from which a PreparedStatement can be created
-     * @param stmtMap the existing map of SQL to PreparedStatements
-     * @param obtainKeys whether or not we need to obtain generated keys for INSERT statements
-     * @param fragmentedTransaction whether or not the SQL pertains to a fragmented transaction
-     *
-     * @return a StatementFlowFileEnclosure to use for executing the given SQL statement
-     *
-     * @throws SQLException if unable to create the appropriate PreparedStatement
-     */
-    private StatementFlowFileEnclosure getEnclosure(final String sql, final Connection conn, final Map<String, StatementFlowFileEnclosure> stmtMap,
-                                                    final boolean obtainKeys, final boolean fragmentedTransaction) throws SQLException {
-        StatementFlowFileEnclosure enclosure = stmtMap.get(sql);
-        if (enclosure != null) {
-            return enclosure;
-        }
-
-        if (obtainKeys) {
-            // Create a new Prepared Statement, requesting that it return the generated keys.
-            PreparedStatement stmt = conn.prepareStatement(sql, Statement.RETURN_GENERATED_KEYS);
-
-            if (stmt == null) {
-                // since we are passing Statement.RETURN_GENERATED_KEYS, calls to conn.prepareStatement will
-                // in some cases (at least for DerbyDB) return null.
-                // We will attempt to recompile the statement without the generated keys being returned.
-                stmt = conn.prepareStatement(sql);
-            }
-
-            // If we need to obtain keys, then we cannot do a Batch Update. In this case,
-            // we don't need to store the PreparedStatement in the Map because we aren't
-            // doing an addBatch/executeBatch. Instead, we will use the statement once
-            // and close it.
-            return new StatementFlowFileEnclosure(stmt);
-        } else if (fragmentedTransaction) {
-            // We cannot use Batch Updates if we have a transaction that spans multiple FlowFiles.
-            // If we did, we could end up processing the statements out of order. It's quite possible
-            // that we could refactor the code some to allow for this, but as it is right now, this
-            // could cause problems. This is because we have a Map<String, StatementFlowFileEnclosure>.
-            // If we had a transaction that needed to execute Stmt A with some parameters, then Stmt B with
-            // some parameters, then Stmt A with different parameters, this would become problematic because
-            // the executeUpdate would be evaluated first for Stmt A (the 1st and 3rd statements, and then
-            // the second statement would be evaluated).
-            final PreparedStatement stmt = conn.prepareStatement(sql);
-            return new StatementFlowFileEnclosure(stmt);
-        }
-
-        final PreparedStatement stmt = conn.prepareStatement(sql);
-        enclosure = new StatementFlowFileEnclosure(stmt);
-        stmtMap.put(sql, enclosure);
-        return enclosure;
-    }
-
-
     /**
      * Determines the SQL statement that should be executed for the given FlowFile
      *
@@ -618,7 +664,7 @@ public class PutSQL extends AbstractProcessor {
 
                 final boolean isNumeric = NUMBER_PATTERN.matcher(entry.getValue()).matches();
                 if (!isNumeric) {
-                    throw new ProcessException("Value of the " + key + " attribute is '" + entry.getValue() + "', which is not a valid JDBC numeral type");
+                    throw new SQLDataException("Value of the " + key + " attribute is '" + entry.getValue() + "', which is not a valid JDBC numeral type");
                 }
 
                 final int jdbcType = Integer.parseInt(entry.getValue());
@@ -630,11 +676,11 @@ public class PutSQL extends AbstractProcessor {
                 try {
                     setParameter(stmt, valueAttrName, parameterIndex, parameterValue, jdbcType, parameterFormat);
                 } catch (final NumberFormatException nfe) {
-                    throw new ProcessException("The value of the " + valueAttrName + " is '" + parameterValue + "', which cannot be converted into the necessary data type", nfe);
+                    throw new SQLDataException("The value of the " + valueAttrName + " is '" + parameterValue + "', which cannot be converted into the necessary data type", nfe);
                 } catch (ParseException pe) {
-                    throw new ProcessException("The value of the " + valueAttrName + " is '" + parameterValue + "', which cannot be converted to a timestamp", pe);
+                    throw new SQLDataException("The value of the " + valueAttrName + " is '" + parameterValue + "', which cannot be converted to a timestamp", pe);
                 } catch (UnsupportedEncodingException uee) {
-                    throw new ProcessException("The value of the " + valueAttrName + " is '" + parameterValue + "', which cannot be converted to UTF-8", uee);
+                    throw new SQLDataException("The value of the " + valueAttrName + " is '" + parameterValue + "', which cannot be converted to UTF-8", uee);
                 }
             }
         }
@@ -652,76 +698,69 @@ public class PutSQL extends AbstractProcessor {
      * @return the appropriate relationship to route the FlowFiles to, or <code>null</code> if the FlowFiles
      *         should instead be processed
      */
-    Relationship determineRelationship(final List<FlowFile> flowFiles, final Long transactionTimeoutMillis) {
+    boolean isFragmentedTransactionReady(final List<FlowFile> flowFiles, final Long transactionTimeoutMillis) throws IllegalArgumentException {
         int selectedNumFragments = 0;
         final BitSet bitSet = new BitSet();
 
+        BiFunction<String, Object[], IllegalArgumentException> illegal = (s, objects) -> new IllegalArgumentException(String.format(s, objects));
+
         for (final FlowFile flowFile : flowFiles) {
             final String fragmentCount = flowFile.getAttribute(FRAGMENT_COUNT_ATTR);
             if (fragmentCount == null && flowFiles.size() == 1) {
-                return null;
+                return true;
             } else if (fragmentCount == null) {
-                getLogger().error("Cannot process {} because there are {} FlowFiles with the same fragment.identifier "
+                throw illegal.apply("Cannot process %s because there are %d FlowFiles with the same fragment.identifier "
-                        + "attribute but not all FlowFiles have a fragment.count attribute; routing all to failure", new Object[] {flowFile, flowFiles.size()});
+                        + "attribute but not all FlowFiles have a fragment.count attribute", new Object[] {flowFile, flowFiles.size()});
-                return REL_FAILURE;
             }
 
             final int numFragments;
             try {
                 numFragments = Integer.parseInt(fragmentCount);
             } catch (final NumberFormatException nfe) {
-                getLogger().error("Cannot process {} because the fragment.count attribute has a value of '{}', which is not an integer; "
+                throw illegal.apply("Cannot process %s because the fragment.count attribute has a value of '%s', which is not an integer",
-                        + "routing all FlowFiles with this fragment.identifier to failure", new Object[] {flowFile, fragmentCount});
+                        new Object[] {flowFile, fragmentCount});
-                return REL_FAILURE;
             }
 
             if (numFragments < 1) {
-                getLogger().error("Cannot process {} because the fragment.count attribute has a value of '{}', which is not a positive integer; "
+                throw illegal.apply("Cannot process %s because the fragment.count attribute has a value of '%s', which is not a positive integer",
-                        + "routing all FlowFiles with this fragment.identifier to failure", new Object[] {flowFile, fragmentCount});
+                        new Object[] {flowFile, fragmentCount});
-                return REL_FAILURE;
             }
 
             if (selectedNumFragments == 0) {
                 selectedNumFragments = numFragments;
             } else if (numFragments != selectedNumFragments) {
-                getLogger().error("Cannot process {} because the fragment.count attribute has different values for different FlowFiles with the same fragment.identifier; "
+                throw illegal.apply("Cannot process %s because the fragment.count attribute has different values for different FlowFiles with the same fragment.identifier",
-                        + "routing all FlowFiles with this fragment.identifier to failure", new Object[] {flowFile});
+                        new Object[] {flowFile});
-                return REL_FAILURE;
             }
 
             final String fragmentIndex = flowFile.getAttribute(FRAGMENT_INDEX_ATTR);
             if (fragmentIndex == null) {
-                getLogger().error("Cannot process {} because the fragment.index attribute is missing; "
+                throw illegal.apply("Cannot process %s because the fragment.index attribute is missing", new Object[] {flowFile});
-                        + "routing all FlowFiles with this fragment.identifier to failure", new Object[] {flowFile});
-                return REL_FAILURE;
             }
 
             final int idx;
             try {
                 idx = Integer.parseInt(fragmentIndex);
             } catch (final NumberFormatException nfe) {
-                getLogger().error("Cannot process {} because the fragment.index attribute has a value of '{}', which is not an integer; "
+                throw illegal.apply("Cannot process %s because the fragment.index attribute has a value of '%s', which is not an integer",
-                        + "routing all FlowFiles with this fragment.identifier to failure", new Object[] {flowFile, fragmentIndex});
+                        new Object[] {flowFile, fragmentIndex});
-                return REL_FAILURE;
             }
 
             if (idx < 0) {
-                getLogger().error("Cannot process {} because the fragment.index attribute has a value of '{}', which is not a positive integer; "
+                throw illegal.apply("Cannot process %s because the fragment.index attribute has a value of '%s', which is not a positive integer",
-                        + "routing all FlowFiles with this fragment.identifier to failure", new Object[] {flowFile, fragmentIndex});
+                        new Object[] {flowFile, fragmentIndex});
-                return REL_FAILURE;
             }
 
             if (bitSet.get(idx)) {
-                getLogger().error("Cannot process {} because it has the same value for the fragment.index attribute as another FlowFile with the same fragment.identifier; "
+                throw illegal.apply("Cannot process %s because it has the same value for the fragment.index attribute as another FlowFile with the same fragment.identifier",
-                        + "routing all FlowFiles with this fragment.identifier to failure", new Object[] {flowFile});
+                        new Object[] {flowFile});
-                return REL_FAILURE;
             }
 
             bitSet.set(idx);
         }
 
         if (selectedNumFragments == flowFiles.size()) {
-            return null; // no relationship to route FlowFiles to yet - process the FlowFiles.
+            return true; // no relationship to route FlowFiles to yet - process the FlowFiles.
         }
 
         long latestQueueTime = 0L;
@@ -733,13 +772,12 @@ public class PutSQL extends AbstractProcessor {
 
         if (transactionTimeoutMillis != null) {
             if (latestQueueTime > 0L && System.currentTimeMillis() - latestQueueTime > transactionTimeoutMillis) {
-                getLogger().error("The transaction timeout has expired for the following FlowFiles; they will be routed to failure: {}", new Object[] {flowFiles});
+                throw illegal.apply("The transaction timeout has expired for the following FlowFiles; they will be routed to failure: %s", new Object[] {flowFiles});
-                return REL_FAILURE;
             }
         }
 
         getLogger().debug("Not enough FlowFiles for transaction. Returning all FlowFiles to queue");
-        return Relationship.SELF; // not enough FlowFiles for this transaction. Return them all to queue.
+        return false;  // not enough FlowFiles for this transaction. Return them all to queue.
     }
 
     /**
@@ -924,7 +962,7 @@ public class PutSQL extends AbstractProcessor {
             if (selectedId.equals(fragmentId)) {
                 // fragment id's match. Find out if we have all of the necessary fragments or not.
                 final int numFragments;
-                if (NUMBER_PATTERN.matcher(fragCount).matches()) {
+                if (fragCount != null && NUMBER_PATTERN.matcher(fragCount).matches()) {
                     numFragments = Integer.parseInt(fragCount);
                 } else {
                     numFragments = Integer.MAX_VALUE;
@@ -971,22 +1009,69 @@ public class PutSQL extends AbstractProcessor {
     }
 
 
+    private static class FragmentedEnclosure extends StatementFlowFileEnclosure {
+
+        private final Map<FlowFile, StatementFlowFileEnclosure> flowFileToEnclosure = new HashMap<>();
+
+        public FragmentedEnclosure() {
+            super(null);
+        }
+
+        public void addFlowFile(final FlowFile flowFile, final StatementFlowFileEnclosure enclosure) {
+            addFlowFile(flowFile);
+            flowFileToEnclosure.put(flowFile, enclosure);
+        }
+
+        public StatementFlowFileEnclosure getTargetEnclosure(final FlowFile flowFile) {
+            return flowFileToEnclosure.get(flowFile);
+        }
+    }
+
     /**
      * A simple, immutable data structure to hold a Prepared Statement and a List of FlowFiles
      * for which that statement should be evaluated.
      */
-    private static class StatementFlowFileEnclosure {
+    private static class StatementFlowFileEnclosure implements FlowFileGroup {
-        private final PreparedStatement statement;
+        private final String sql;
+        private PreparedStatement statement;
         private final List<FlowFile> flowFiles = new ArrayList<>();
 
-        public StatementFlowFileEnclosure(final PreparedStatement statement) {
+        public StatementFlowFileEnclosure(String sql) {
-            this.statement = statement;
+            this.sql = sql;
         }
 
-        public PreparedStatement getStatement() {
+        public PreparedStatement getNewStatement(final Connection conn, final boolean obtainKeys) throws SQLException {
+            if (obtainKeys) {
+                // Create a new Prepared Statement, requesting that it return the generated keys.
+                PreparedStatement stmt = conn.prepareStatement(sql, Statement.RETURN_GENERATED_KEYS);
+
+                if (stmt == null) {
+                    // since we are passing Statement.RETURN_GENERATED_KEYS, calls to conn.prepareStatement will
+                    // in some cases (at least for DerbyDB) return null.
+                    // We will attempt to recompile the statement without the generated keys being returned.
+                    stmt = conn.prepareStatement(sql);
+                }
+
+                // If we need to obtain keys, then we cannot do a Batch Update. In this case,
+                // we don't need to store the PreparedStatement in the Map because we aren't
+                // doing an addBatch/executeBatch. Instead, we will use the statement once
+                // and close it.
+                return stmt;
+            }
+
+            return conn.prepareStatement(sql);
+        }
+
+        public PreparedStatement getCachedStatement(final Connection conn) throws SQLException {
+            if (statement != null) {
+                return statement;
+            }
+
+            statement = conn.prepareStatement(sql);
             return statement;
         }
 
+        @Override
         public List<FlowFile> getFlowFiles() {
             return flowFiles;
         }
@@ -997,7 +1082,7 @@ public class PutSQL extends AbstractProcessor {
 
         @Override
         public int hashCode() {
-            return statement.hashCode();
+            return sql.hashCode();
         }
 
         @Override
@@ -1013,7 +1098,7 @@ public class PutSQL extends AbstractProcessor {
             }
 
             final StatementFlowFileEnclosure other = (StatementFlowFileEnclosure) obj;
-            return statement.equals(other.getStatement());
+            return sql.equals(other.sql);
         }
     }
 }

nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/test/groovy/org/apache/nifi/processors/standard/TestPutDatabaseRecord.groovy

@@ -17,6 +17,7 @@
 package org.apache.nifi.processors.standard
 
 import org.apache.nifi.processor.exception.ProcessException
+import org.apache.nifi.processor.util.pattern.RollbackOnFailure
 import org.apache.nifi.processors.standard.util.record.MockRecordParser
 import org.apache.nifi.reporting.InitializationException
 import org.apache.nifi.serialization.record.RecordField
@@ -42,7 +43,6 @@ import java.sql.Statement
 
 import static org.junit.Assert.assertEquals
 import static org.junit.Assert.assertFalse
-import static org.junit.Assert.assertNull
 import static org.junit.Assert.assertTrue
 import static org.junit.Assert.fail
 import static org.mockito.Mockito.spy
@@ -53,7 +53,8 @@ import static org.mockito.Mockito.spy
 @RunWith(JUnit4.class)
 class TestPutDatabaseRecord {
 
-    private static final String createPersons = "CREATE TABLE PERSONS (id integer primary key, name varchar(100), code integer)"
+    private static final String createPersons = "CREATE TABLE PERSONS (id integer primary key, name varchar(100)," +
+            " code integer CONSTRAINT CODE_RANGE CHECK (code >= 0 AND code < 1000))"
     private final static String DB_LOCATION = "target/db_pdr"
 
     TestRunner runner
@@ -131,33 +132,20 @@ class TestPutDatabaseRecord {
 
         ] as PutDatabaseRecord.TableSchema
 
+        runner.setProperty(PutDatabaseRecord.TRANSLATE_FIELD_NAMES, 'false')
+        runner.setProperty(PutDatabaseRecord.UNMATCHED_FIELD_BEHAVIOR, 'false')
+        runner.setProperty(PutDatabaseRecord.UNMATCHED_COLUMN_BEHAVIOR, 'false')
+        runner.setProperty(PutDatabaseRecord.QUOTED_IDENTIFIERS, 'false')
+        runner.setProperty(PutDatabaseRecord.QUOTED_TABLE_IDENTIFIER, 'false')
+        def settings = new PutDatabaseRecord.DMLSettings(runner.getProcessContext())
+
         processor.with {
-            try {
-                assertNull(generateInsert(null, null, null,
-                        false, false, false, false,
-                        false, false).sql)
-                fail('Expecting ProcessException')
-            } catch (ProcessException ignore) {
-                // Expected
-            }
-            try {
-                assertNull(generateInsert(null, 'PERSONS', null,
-                        false, false, false, false,
-                        false, false).sql)
-                fail('Expecting ProcessException')
-            } catch (ProcessException ignore) {
-                // Expected
-            }
 
             assertEquals('INSERT INTO PERSONS (id, name, code) VALUES (?,?,?)',
-                    generateInsert(schema, 'PERSONS', tableSchema,
+                    generateInsert(schema, 'PERSONS', tableSchema, settings).sql)
-                            false, false, false, false,
-                            false, false).sql)
 
             assertEquals('DELETE FROM PERSONS WHERE id = ? AND name = ? AND code = ?',
-                    generateDelete(schema, 'PERSONS', tableSchema,
+                    generateDelete(schema, 'PERSONS', tableSchema, settings).sql)
-                            false, false, false, false,
-                            false, false).sql)
         }
     }
 
@@ -210,6 +198,81 @@ class TestPutDatabaseRecord {
         conn.close()
     }
 
+    @Test
+    void testInsertBatchUpdateException() throws InitializationException, ProcessException, SQLException, IOException {
+        recreateTable("PERSONS", createPersons)
+        final MockRecordParser parser = new MockRecordParser()
+        runner.addControllerService("parser", parser)
+        runner.enableControllerService(parser)
+
+        parser.addSchemaField("id", RecordFieldType.INT)
+        parser.addSchemaField("name", RecordFieldType.STRING)
+        parser.addSchemaField("code", RecordFieldType.INT)
+
+        parser.addRecord(1, 'rec1', 101)
+        parser.addRecord(2, 'rec2', 102)
+        parser.addRecord(3, 'rec3', 1000)
+        parser.addRecord(4, 'rec4', 104)
+
+        runner.setProperty(PutDatabaseRecord.RECORD_READER_FACTORY, 'parser')
+        runner.setProperty(PutDatabaseRecord.STATEMENT_TYPE, PutDatabaseRecord.INSERT_TYPE)
+        runner.setProperty(PutDatabaseRecord.TABLE_NAME, 'PERSONS')
+
+        runner.enqueue(new byte[0])
+        runner.run()
+
+        runner.assertTransferCount(PutDatabaseRecord.REL_SUCCESS, 0)
+        runner.assertTransferCount(PutDatabaseRecord.REL_FAILURE, 0)
+        runner.assertTransferCount(PutDatabaseRecord.REL_RETRY, 1)
+        final Connection conn = dbcp.getConnection()
+        final Statement stmt = conn.createStatement()
+        final ResultSet rs = stmt.executeQuery('SELECT * FROM PERSONS')
+        // Transaction should be rolled back and table should remain empty.
+        assertFalse(rs.next())
+
+        stmt.close()
+        conn.close()
+    }
+
+    @Test
+    void testInsertBatchUpdateExceptionRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        recreateTable("PERSONS", createPersons)
+        final MockRecordParser parser = new MockRecordParser()
+        runner.addControllerService("parser", parser)
+        runner.enableControllerService(parser)
+
+        parser.addSchemaField("id", RecordFieldType.INT)
+        parser.addSchemaField("name", RecordFieldType.STRING)
+        parser.addSchemaField("code", RecordFieldType.INT)
+
+        parser.addRecord(1, 'rec1', 101)
+        parser.addRecord(2, 'rec2', 102)
+        parser.addRecord(3, 'rec3', 1000)
+        parser.addRecord(4, 'rec4', 104)
+
+        runner.setProperty(PutDatabaseRecord.RECORD_READER_FACTORY, 'parser')
+        runner.setProperty(PutDatabaseRecord.STATEMENT_TYPE, PutDatabaseRecord.INSERT_TYPE)
+        runner.setProperty(PutDatabaseRecord.TABLE_NAME, 'PERSONS')
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, 'true')
+
+        runner.enqueue(new byte[0])
+        try {
+            runner.run()
+            fail("ProcessException is expected")
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException)
+        }
+
+        final Connection conn = dbcp.getConnection()
+        final Statement stmt = conn.createStatement()
+        final ResultSet rs = stmt.executeQuery('SELECT * FROM PERSONS')
+        // Transaction should be rolled back and table should remain empty.
+        assertFalse(rs.next())
+
+        stmt.close()
+        conn.close()
+    }
+
     @Test
     void testInsertNoTable() throws InitializationException, ProcessException, SQLException, IOException {
         recreateTable("PERSONS", createPersons)
@@ -299,6 +362,37 @@ class TestPutDatabaseRecord {
         runner.assertTransferCount(PutDatabaseRecord.REL_FAILURE, 1)
     }
 
+    @Test
+    void testSqlStatementTypeNoValueRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        recreateTable("PERSONS", createPersons)
+        final MockRecordParser parser = new MockRecordParser()
+        runner.addControllerService("parser", parser)
+        runner.enableControllerService(parser)
+
+        parser.addSchemaField("sql", RecordFieldType.STRING)
+
+        parser.addRecord('')
+
+        runner.setProperty(PutDatabaseRecord.RECORD_READER_FACTORY, 'parser')
+        runner.setProperty(PutDatabaseRecord.STATEMENT_TYPE, PutDatabaseRecord.USE_ATTR_TYPE)
+        runner.setProperty(PutDatabaseRecord.TABLE_NAME, 'PERSONS')
+        runner.setProperty(PutDatabaseRecord.FIELD_CONTAINING_SQL, 'sql')
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, 'true')
+
+        def attrs = [:]
+        attrs[PutDatabaseRecord.STATEMENT_TYPE_ATTRIBUTE] = 'sql'
+        runner.enqueue(new byte[0], attrs)
+        try {
+            runner.run()
+            fail("ProcessException is expected")
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException)
+        }
+
+        runner.assertTransferCount(PutDatabaseRecord.REL_SUCCESS, 0)
+        runner.assertTransferCount(PutDatabaseRecord.REL_FAILURE, 0)
+    }
+
     @Test
     void testUpdate() throws InitializationException, ProcessException, SQLException, IOException {
         recreateTable("PERSONS", createPersons)

nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/test/java/org/apache/nifi/processors/standard/TestPutSQL.java

@@ -19,6 +19,7 @@ package org.apache.nifi.processors.standard;
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertTrue;
+import static org.junit.Assert.fail;
 
 import java.io.File;
 import java.io.IOException;
@@ -39,6 +40,7 @@ import org.apache.commons.lang3.RandomUtils;
 import org.apache.nifi.controller.AbstractControllerService;
 import org.apache.nifi.dbcp.DBCPService;
 import org.apache.nifi.processor.exception.ProcessException;
+import org.apache.nifi.processor.util.pattern.RollbackOnFailure;
 import org.apache.nifi.reporting.InitializationException;
 import org.apache.nifi.util.MockFlowFile;
 import org.apache.nifi.util.TestRunner;
@@ -165,6 +167,29 @@ public class TestPutSQL {
         runner.assertTransferCount(PutSQL.REL_SUCCESS, 3);
     }
 
+    @Test
+    public void testFailInMiddleWithBadStatementRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(PutSQL.OBTAIN_GENERATED_KEYS, "false");
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+        runner.enqueue("INSERT INTO PERSONS_AI (NAME, CODE) VALUES ('Mark', 84)".getBytes());
+        runner.enqueue("INSERT INTO PERSONS_AI".getBytes()); // intentionally wrong syntax
+        runner.enqueue("INSERT INTO PERSONS_AI (NAME, CODE) VALUES ('Tom', 3)".getBytes());
+        runner.enqueue("INSERT INTO PERSONS_AI (NAME, CODE) VALUES ('Harry', 44)".getBytes());
+
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+            runner.assertTransferCount(PutSQL.REL_FAILURE, 0);
+            runner.assertTransferCount(PutSQL.REL_SUCCESS, 0);
+        }
+    }
+
 
     @Test
     public void testFailInMiddleWithBadParameterType() throws InitializationException, ProcessException, SQLException, IOException {
@@ -191,8 +216,41 @@ public class TestPutSQL {
 
         runner.assertTransferCount(PutSQL.REL_FAILURE, 1);
         runner.assertTransferCount(PutSQL.REL_SUCCESS, 3);
+
     }
 
+    @Test
+    public void testFailInMiddleWithBadParameterTypeRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(PutSQL.OBTAIN_GENERATED_KEYS, "false");
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        final Map<String, String> goodAttributes = new HashMap<>();
+        goodAttributes.put("sql.args.1.type", String.valueOf(Types.INTEGER));
+        goodAttributes.put("sql.args.1.value", "84");
+
+        final Map<String, String> badAttributes = new HashMap<>();
+        badAttributes.put("sql.args.1.type", String.valueOf(Types.VARCHAR));
+        badAttributes.put("sql.args.1.value", "hello");
+
+        final byte[] data = "INSERT INTO PERSONS_AI (NAME, CODE) VALUES ('Mark', ?)".getBytes();
+        runner.enqueue(data, goodAttributes);
+        runner.enqueue(data, badAttributes);
+        runner.enqueue(data, goodAttributes);
+        runner.enqueue(data, goodAttributes);
+
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+            runner.assertTransferCount(PutSQL.REL_FAILURE, 0);
+            runner.assertTransferCount(PutSQL.REL_SUCCESS, 0);
+        }
+    }
 
     @Test
     public void testFailInMiddleWithBadParameterValue() throws InitializationException, ProcessException, SQLException, IOException {
@@ -235,6 +293,48 @@ public class TestPutSQL {
         }
     }
 
+    @Test
+    public void testFailInMiddleWithBadParameterValueRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(PutSQL.OBTAIN_GENERATED_KEYS, "false");
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        recreateTable("PERSONS_AI",createPersonsAutoId);
+
+        final Map<String, String> goodAttributes = new HashMap<>();
+        goodAttributes.put("sql.args.1.type", String.valueOf(Types.INTEGER));
+        goodAttributes.put("sql.args.1.value", "84");
+
+        final Map<String, String> badAttributes = new HashMap<>();
+        badAttributes.put("sql.args.1.type", String.valueOf(Types.INTEGER));
+        badAttributes.put("sql.args.1.value", "9999");
+
+        final byte[] data = "INSERT INTO PERSONS_AI (NAME, CODE) VALUES ('Mark', ?)".getBytes();
+        runner.enqueue(data, goodAttributes);
+        runner.enqueue(data, badAttributes);
+        runner.enqueue(data, goodAttributes);
+        runner.enqueue(data, goodAttributes);
+
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+            runner.assertTransferCount(PutSQL.REL_FAILURE, 0);
+            runner.assertTransferCount(PutSQL.REL_SUCCESS, 0);
+        }
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                final ResultSet rs = stmt.executeQuery("SELECT * FROM PERSONS_AI");
+                assertFalse(rs.next());
+            }
+        }
+    }
+
 
     @Test
     public void testUsingSqlDataTypesWithNegativeValues() throws InitializationException, ProcessException, SQLException, IOException {
@@ -666,6 +766,8 @@ public class TestPutSQL {
         runner.enableControllerService(service);
         runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
 
+        recreateTable("PERSONS", createPersons);
+
         final String sql = "INSERT INTO PERSONS (ID, NAME, CODE) VALUES (?, ?, ?); " +
                 "UPDATE PERSONS SET NAME='George' WHERE ID=?; ";
         final Map<String, String> attributes = new HashMap<>();
@@ -695,6 +797,47 @@ public class TestPutSQL {
         }
     }
 
+    @Test
+    public void testMultipleStatementsWithinFlowFileRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        recreateTable("PERSONS", createPersons);
+
+        final String sql = "INSERT INTO PERSONS (ID, NAME, CODE) VALUES (?, ?, ?); " +
+                "UPDATE PERSONS SET NAME='George' WHERE ID=?; ";
+        final Map<String, String> attributes = new HashMap<>();
+        attributes.put("sql.args.1.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.1.value", "1");
+
+        attributes.put("sql.args.2.type", String.valueOf(Types.VARCHAR));
+        attributes.put("sql.args.2.value", "Mark");
+
+        attributes.put("sql.args.3.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.3.value", "84");
+
+        attributes.put("sql.args.4.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.4.value", "1");
+
+        runner.enqueue(sql.getBytes(), attributes);
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                final ResultSet rs = stmt.executeQuery("SELECT * FROM PERSONS");
+                assertFalse(rs.next());
+            }
+        }
+    }
+
 
     @Test
     public void testWithNullParameter() throws InitializationException, ProcessException, SQLException, IOException {
@@ -766,6 +909,47 @@ public class TestPutSQL {
         }
     }
 
+    @Test
+    public void testInvalidStatementRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        recreateTable("PERSONS", createPersons);
+
+        final String sql = "INSERT INTO PERSONS (ID, NAME, CODE) VALUES (?, ?, ?); " +
+                "UPDATE SOME_RANDOM_TABLE NAME='George' WHERE ID=?; ";
+        final Map<String, String> attributes = new HashMap<>();
+        attributes.put("sql.args.1.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.1.value", "1");
+
+        attributes.put("sql.args.2.type", String.valueOf(Types.VARCHAR));
+        attributes.put("sql.args.2.value", "Mark");
+
+        attributes.put("sql.args.3.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.3.value", "84");
+
+        attributes.put("sql.args.4.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.4.value", "1");
+
+        runner.enqueue(sql.getBytes(), attributes);
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        try (final Connection conn = service.getConnection()) {
+            try (final Statement stmt = conn.createStatement()) {
+                final ResultSet rs = stmt.executeQuery("SELECT * FROM PERSONS");
+                assertFalse(rs.next());
+            }
+        }
+    }
+
 
     @Test
     public void testRetryableFailure() throws InitializationException, ProcessException, SQLException, IOException {
@@ -798,6 +982,42 @@ public class TestPutSQL {
         runner.assertAllFlowFilesTransferred(PutSQL.REL_RETRY, 1);
     }
 
+    @Test
+    public void testRetryableFailureRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        final DBCPService service = new SQLExceptionService(null);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        final String sql = "INSERT INTO PERSONS (ID, NAME, CODE) VALUES (?, ?, ?); " +
+                "UPDATE PERSONS SET NAME='George' WHERE ID=?; ";
+        final Map<String, String> attributes = new HashMap<>();
+        attributes.put("sql.args.1.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.1.value", "1");
+
+        attributes.put("sql.args.2.type", String.valueOf(Types.VARCHAR));
+        attributes.put("sql.args.2.value", "Mark");
+
+        attributes.put("sql.args.3.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.3.value", "84");
+
+        attributes.put("sql.args.4.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.4.value", "1");
+
+        runner.enqueue(sql.getBytes(), attributes);
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+            // Should not be routed to retry.
+            runner.assertAllFlowFilesTransferred(PutSQL.REL_RETRY, 0);
+        }
+
+    }
 
     @Test
     public void testMultipleFlowFilesSuccessfulInTransaction() throws InitializationException, ProcessException, SQLException, IOException {
@@ -857,6 +1077,38 @@ public class TestPutSQL {
         }
     }
 
+    @Test
+    public void testMultipleFlowFilesSuccessfulInTransactionRollBackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(PutSQL.BATCH_SIZE, "1");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+
+        recreateTable("PERSONS", createPersons);
+
+        final Map<String, String> attributes = new HashMap<>();
+        attributes.put("sql.args.1.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.1.value", "1");
+
+        attributes.put("sql.args.2.type", String.valueOf(Types.VARCHAR));
+        attributes.put("sql.args.2.value", "Mark");
+
+        attributes.put("sql.args.3.type", String.valueOf(Types.INTEGER));
+        attributes.put("sql.args.3.value", "84");
+
+        attributes.put("fragment.identifier", "1");
+        attributes.put("fragment.count", "2");
+        attributes.put("fragment.index", "0");
+        runner.enqueue("INSERT INTO PERSONS (ID, NAME, CODE) VALUES (?, ?, ?)".getBytes(), attributes);
+        // ProcessException should not be thrown in this case, because the input FlowFiles are simply differed.
+        runner.run();
+
+        // No FlowFiles should be transferred because there were not enough flowfiles with the same fragment identifier
+        runner.assertAllFlowFilesTransferred(PutSQL.REL_SUCCESS, 0);
+
+    }
 
     @Test
     public void testTransactionTimeout() throws InitializationException, ProcessException, SQLException, IOException {
@@ -895,6 +1147,81 @@ public class TestPutSQL {
         runner.assertAllFlowFilesTransferred(PutSQL.REL_FAILURE, 1);
     }
 
+    @Test
+    public void testTransactionTimeoutRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+
+        runner.setProperty(PutSQL.TRANSACTION_TIMEOUT, "5 secs");
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+        final Map<String, String> attributes = new HashMap<>();
+        attributes.put("fragment.identifier", "1");
+        attributes.put("fragment.count", "2");
+        attributes.put("fragment.index", "0");
+
+        final MockFlowFile mff = new MockFlowFile(0L) {
+            @Override
+            public Long getLastQueueDate() {
+                return System.currentTimeMillis() - 10000L; // return 10 seconds ago
+            }
+
+            @Override
+            public Map<String, String> getAttributes() {
+                return attributes;
+            }
+
+            @Override
+            public String getAttribute(final String attrName) {
+                return attributes.get(attrName);
+            }
+        };
+
+        runner.enqueue(mff);
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertAllFlowFilesTransferred(PutSQL.REL_FAILURE, 0);
+    }
+
+    @Test
+    public void testNullFragmentCountRollbackOnFailure() throws InitializationException, ProcessException, SQLException, IOException {
+        final TestRunner runner = TestRunners.newTestRunner(PutSQL.class);
+        runner.addControllerService("dbcp", service);
+        runner.enableControllerService(service);
+
+        runner.setProperty(PutSQL.TRANSACTION_TIMEOUT, "5 secs");
+        runner.setProperty(PutSQL.CONNECTION_POOL, "dbcp");
+        runner.setProperty(RollbackOnFailure.ROLLBACK_ON_FAILURE, "true");
+        final Map<String, String> attribute1 = new HashMap<>();
+        attribute1.put("fragment.identifier", "1");
+        attribute1.put("fragment.count", "2");
+        attribute1.put("fragment.index", "0");
+
+        final Map<String, String> attribute2 = new HashMap<>();
+        attribute2.put("fragment.identifier", "1");
+//        attribute2.put("fragment.count", null);
+        attribute2.put("fragment.index", "1");
+
+        runner.enqueue(new byte[]{}, attribute1);
+        runner.enqueue(new byte[]{}, attribute2);
+
+
+        try {
+            runner.run();
+            fail("ProcessException should be thrown");
+        } catch (AssertionError e) {
+            assertTrue(e.getCause() instanceof ProcessException);
+        }
+
+        runner.assertAllFlowFilesTransferred(PutSQL.REL_FAILURE, 0);
+    }
+
     /**
      * Simple implementation only for testing purposes
      */
@@ -985,4 +1312,5 @@ public class TestPutSQL {
         byte[] bBinary = RandomUtils.nextBytes(length);
         return DatatypeConverter.printBase64Binary(bBinary);
     }
+
 }