[ https://issues.apache.org/jira/browse/NIFI-3792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16655409#comment-16655409 ]
ASF GitHub Bot commented on NIFI-3792: -------------------------------------- Github user patricker commented on a diff in the pull request: https://github.com/apache/nifi/pull/3090#discussion_r226347867 --- Diff: nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/RetryCount.java --- @@ -0,0 +1,210 @@ +/* + * 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.processors.standard; + +import org.apache.nifi.annotation.behavior.InputRequirement; +import org.apache.nifi.annotation.behavior.ReadsAttribute; +import org.apache.nifi.annotation.behavior.SideEffectFree; +import org.apache.nifi.annotation.behavior.SupportsBatching; +import org.apache.nifi.annotation.behavior.WritesAttribute; +import org.apache.nifi.annotation.documentation.CapabilityDescription; +import org.apache.nifi.annotation.documentation.Tags; +import org.apache.nifi.annotation.lifecycle.OnScheduled; +import org.apache.nifi.components.PropertyDescriptor; +import org.apache.nifi.flowfile.FlowFile; +import org.apache.nifi.processor.AbstractProcessor; +import org.apache.nifi.processor.DataUnit; +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.StandardValidators; + +import java.util.ArrayList; +import java.util.Collections; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; + +import static org.apache.nifi.processors.standard.RetryCount.RETRY_COUNT_ATTRIBUTE_KEY; +import static org.apache.nifi.annotation.behavior.InputRequirement.Requirement.INPUT_REQUIRED; +import static org.apache.nifi.expression.ExpressionLanguageScope.VARIABLE_REGISTRY; + +/** + * Created by jpercivall on 3/17/17. + */ +@SupportsBatching +@SideEffectFree +@InputRequirement(INPUT_REQUIRED) +@Tags({"penalize", "retry", "penalize"}) +@CapabilityDescription("Used to facilitate retrying a FlowFile a set number of times before considering the FlowFile 'failed'. Can also be used as a utility to penalize FlowFiles.") +@WritesAttribute(attribute = RETRY_COUNT_ATTRIBUTE_KEY, description = "The number of times this FlowFile has been routed to 'Retry'.") +@ReadsAttribute(attribute = RETRY_COUNT_ATTRIBUTE_KEY, description = "The number of times this FlowFile has been routed to 'Retry'.") +public class RetryCount extends AbstractProcessor { + + public static final String RETRY_COUNT_ATTRIBUTE_KEY = "retryCountProcessor.timesRetried"; + + public static final PropertyDescriptor PROP_RETRY_LIMIT = new PropertyDescriptor.Builder() + .name("Retry limit") + .displayName("Retry limit") + .description("The number of times to retry the FlowFile before considering it 'failed' and routing it to 'over limit'.") + .expressionLanguageSupported(VARIABLE_REGISTRY) + .required(true) + .addValidator(StandardValidators.INTEGER_VALIDATOR) + .defaultValue("3") + .build(); + + public static final PropertyDescriptor PROP_PENALIZE_FLOWFILE = new PropertyDescriptor.Builder() + .name("Penalize FlowFile") + .displayName("Penalize FlowFile") + .description("If true then the FlowFiles routed to 'retry' will be penalized.") + .allowableValues("true", "false") + .defaultValue("true") + .build(); + + public static final PropertyDescriptor PROP_WARN_ON_OVER_LIMIT = new PropertyDescriptor.Builder() + .name("Warn on 'over limit'") + .displayName("Warn on 'over limit'") + .description("If true then when a FlowFile is routed to 'over limit' a message will be logged at the level 'warn'.") + .allowableValues("true", "false") + .defaultValue("true") + .build(); + + public static final PropertyDescriptor PROP_INPUT_QUEUE_DATA_SIZE_LIMIT = new PropertyDescriptor.Builder() + .name("Input queue data size limit") + .displayName("Input queue data size limit") + .description("If the total input queue data size (a combination of all connections feeding this processor) is above this configured limit then any processed FlowFile is routed to " + + "'over limit'. This is necessary to avoid a back-pressure deadlock caused by with circular back-pressure. The default values are set to be half of what the default " + + "configuration values are for connections back-pressure.") + .addValidator(StandardValidators.DATA_SIZE_VALIDATOR) + .expressionLanguageSupported(VARIABLE_REGISTRY) + .defaultValue("500 mb") + .build(); + + public static final PropertyDescriptor PROP_INPUT_QUEUE_COUNT_LIMIT = new PropertyDescriptor.Builder() + .name("Input queue count limit") + .displayName("Input queue count limit") + .description("If the total input queue FlowFile count (a combination of all connections feeding this processor) is above this configured limit then any processed FlowFile is routed to " + + "'over limit'. This is necessary to avoid a back-pressure deadlock caused by with circular back-pressure. The default values are set to be half of what the default " + + "configuration values are for connections back-pressure.") + .addValidator(StandardValidators.POSITIVE_INTEGER_VALIDATOR) + .expressionLanguageSupported(VARIABLE_REGISTRY) + .defaultValue("5000") + .build(); + + + public static final Relationship REL_RETRY = new Relationship.Builder() + .name("retry") + .description("FlowFiles that have not yet exceeded the retry limit are routed here.") + .build(); + + public static final Relationship REL_OVER_LIMIT = new Relationship.Builder() + .name("over limit") + .description("A FlowFile which has exceeded the limit for the number of retries is routed here.") + .build(); + + + private static final List<PropertyDescriptor> descriptors; + private static final Set<Relationship> relationships; + + static { + final List<PropertyDescriptor> descriptorsTemp = new ArrayList<PropertyDescriptor>(); + descriptorsTemp.add(PROP_RETRY_LIMIT); + descriptorsTemp.add(PROP_PENALIZE_FLOWFILE); + descriptorsTemp.add(PROP_WARN_ON_OVER_LIMIT); + descriptorsTemp.add(PROP_INPUT_QUEUE_DATA_SIZE_LIMIT); + descriptorsTemp.add(PROP_INPUT_QUEUE_COUNT_LIMIT); + descriptors = Collections.unmodifiableList(descriptorsTemp); + + final Set<Relationship> relationshipsTemp = new HashSet<Relationship>(); + relationshipsTemp.add(REL_OVER_LIMIT); + relationshipsTemp.add(REL_RETRY); + relationships = relationshipsTemp; + } + + private volatile boolean penalize; + private volatile boolean warnOnOverLimit; + private volatile int retryLimit; + private volatile double dataSizeLimit; + private volatile int objectCountLimit; + + @Override + public Set<Relationship> getRelationships() { + return relationships; + } + + @Override + public final List<PropertyDescriptor> getSupportedPropertyDescriptors() { + return descriptors; + } + + @OnScheduled + public void setPropertyConstants(final ProcessContext context) { + penalize = context.getProperty(PROP_PENALIZE_FLOWFILE).asBoolean(); + warnOnOverLimit = context.getProperty(PROP_WARN_ON_OVER_LIMIT).asBoolean(); + retryLimit = context.getProperty(PROP_RETRY_LIMIT).evaluateAttributeExpressions().asInteger(); + dataSizeLimit = context.getProperty(PROP_INPUT_QUEUE_DATA_SIZE_LIMIT).evaluateAttributeExpressions().asDataSize(DataUnit.B); + objectCountLimit = context.getProperty(PROP_INPUT_QUEUE_COUNT_LIMIT).evaluateAttributeExpressions().asInteger(); + } + + @Override + public void onTrigger(ProcessContext context, ProcessSession session) throws ProcessException { + + FlowFile flowFile = session.get(); + if (flowFile == null) { + return; + } + + final Optional<String> timesRetriedStringOptional = Optional.ofNullable(flowFile.getAttribute(RETRY_COUNT_ATTRIBUTE_KEY)); + + final int timesRetried = Integer.parseInt(timesRetriedStringOptional.orElse("0")); + + if (timesRetried >= retryLimit) { + + if (warnOnOverLimit) { + getLogger().warn("The FlowFile {} has been retried {} times and will be routed to 'over limit'", new Object[]{flowFile, timesRetried}); + } + session.transfer(flowFile, REL_OVER_LIMIT); + + } else { + + int inputObjectCount = session.getQueueSize().getObjectCount(); --- End diff -- You don't have these two props marked as required in the property definitions, but it looks like in the code they are required right now? > A processor to facilitate retrying FlowFiles > -------------------------------------------- > > Key: NIFI-3792 > URL: https://issues.apache.org/jira/browse/NIFI-3792 > Project: Apache NiFi > Issue Type: New Feature > Reporter: Joseph Percivall > Assignee: Joseph Percivall > Priority: Major > > When dealing with processor failures in production, potentially related to > network issues, the current best practice is to retry the flowfile a couple > times before declaring it a failure[1]. This currently requires multiple > processors and penalizing the flowfile isn't possible. Also if the flow is > not fast enough, back-pressure can cause a livelocked state which requires > manual intervention. > A new processor should be added to not only encourage best practices but also > offer configuration options to deal with un-optimal situations. > [1] > https://community.hortonworks.com/questions/77336/nifi-best-practices-for-error-handling.html -- This message was sent by Atlassian JIRA (v7.6.3#76005)