Skip to content

move statemgr construction to streamtask #67

New issue

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

Sign up for GitHub

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

Already on GitHub? Sign in to your account

Merged
ymatsuda merged 1 commit into from
Sep 22, 2015
Merged

move statemgr construction to streamtask #67

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

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 3 additions & 24 deletions streams/src/main/java/org/apache/kafka/streams/processor/internals/ProcessorContextImpl.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,12 +17,9 @@

package org.apache.kafka.streams.processor.internals;

import org.apache.kafka.clients.consumer.Consumer;
import org.apache.kafka.clients.consumer.KafkaConsumer;
import org.apache.kafka.common.KafkaException;
import org.apache.kafka.common.TopicPartition;
import org.apache.kafka.common.metrics.Metrics;
import org.apache.kafka.common.serialization.ByteArrayDeserializer;
import org.apache.kafka.common.serialization.Deserializer;
import org.apache.kafka.common.serialization.Serializer;
import org.apache.kafka.streams.StreamingConfig;
Expand All @@ -34,7 +31,6 @@
import org.slf4j.LoggerFactory;

import java.io.File;
import java.io.IOException;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
Expand Down Expand Up @@ -63,43 +59,26 @@ public ProcessorContextImpl(int id,
StreamTask task,
StreamingConfig config,
RecordCollector collector,
Metrics metrics) throws IOException {
ProcessorStateManager stateMgr,
Metrics metrics) {
this.id = id;
this.task = task;
this.metrics = metrics;
this.collector = collector;
this.stateMgr = stateMgr;

this.keySerializer = config.getConfiguredInstance(StreamingConfig.KEY_SERIALIZER_CLASS_CONFIG, Serializer.class);
this.valSerializer = config.getConfiguredInstance(StreamingConfig.VALUE_SERIALIZER_CLASS_CONFIG, Serializer.class);
this.keyDeserializer = config.getConfiguredInstance(StreamingConfig.KEY_DESERIALIZER_CLASS_CONFIG, Deserializer.class);
this.valDeserializer = config.getConfiguredInstance(StreamingConfig.VALUE_DESERIALIZER_CLASS_CONFIG, Deserializer.class);

File stateFile = new File(config.getString(StreamingConfig.STATE_DIR_CONFIG), Integer.toString(id));

log.info("Creating restoration consumer client for stream task [" + task.id() + "]");

Consumer restoreConsumer = new KafkaConsumer<>(
config.getConsumerConfigs(),
new ByteArrayDeserializer(),
new ByteArrayDeserializer());

this.stateMgr = new ProcessorStateManager(id, stateFile, restoreConsumer);

this.initialized = false;
}

public ProcessorStateManager stateManager() {
return this.stateMgr;
}

public RecordCollector recordCollector() {
return this.collector;
}

public StreamTask task() {
return this.task;
}

@Override
public boolean joinable() {
Set<TopicPartition> partitions = this.task.partitions();
Expand Down
25 changes: 20 additions & 5 deletions streams/src/main/java/org/apache/kafka/streams/processor/internals/StreamTask.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,16 +19,19 @@

import org.apache.kafka.clients.consumer.Consumer;
import org.apache.kafka.clients.consumer.ConsumerRecord;
import org.apache.kafka.clients.consumer.KafkaConsumer;
import org.apache.kafka.clients.producer.Producer;
import org.apache.kafka.common.KafkaException;
import org.apache.kafka.common.TopicPartition;
import org.apache.kafka.common.metrics.Metrics;
import org.apache.kafka.common.serialization.ByteArrayDeserializer;
import org.apache.kafka.streams.StreamingConfig;
import org.apache.kafka.streams.processor.ProcessorContext;
import org.apache.kafka.streams.processor.TimestampExtractor;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import java.io.File;
import java.io.IOException;
import java.util.Collection;
import java.util.HashMap;
Expand All @@ -55,6 +58,7 @@ public class StreamTask implements Punctuator {

private final Map<TopicPartition, Long> consumedOffsets;
private final RecordCollector recordCollector;
private final ProcessorStateManager stateMgr;

private boolean commitRequested = false;
private boolean commitOffsetNeeded = false;
Expand Down Expand Up @@ -104,13 +108,24 @@ public StreamTask(int id,
// create the record recordCollector that maintains the produced offsets
this.recordCollector = new RecordCollector(producer);

// initialize the topology with its own context
log.info("Creating restoration consumer client for stream task [" + id + "]");

Consumer restoreConsumer = new KafkaConsumer<>(
config.getConsumerConfigs(),
new ByteArrayDeserializer(),
new ByteArrayDeserializer());

// create the processor state manager
try {
this.processorContext = new ProcessorContextImpl(id, this, config, recordCollector, new Metrics());
File stateFile = new File(config.getString(StreamingConfig.STATE_DIR_CONFIG), Integer.toString(id));
this.stateMgr = new ProcessorStateManager(id, stateFile, restoreConsumer);
} catch (IOException e) {
throw new KafkaException("Error while creating the state manager in processor context", e);
throw new KafkaException("Error while creating the state manager", e);
}

// initialize the topology with its own context
this.processorContext = new ProcessorContextImpl(id, this, config, recordCollector, stateMgr, new Metrics());

// initialize the task by initializing all its processor nodes in the topology
for (ProcessorNode node : this.topology.processors()) {
this.currNode = node;
Expand Down Expand Up @@ -232,7 +247,7 @@ public ProcessorTopology topology() {
*/
public void commit() {
// 1) flush local state
((ProcessorContextImpl) processorContext).stateManager().flush();
stateMgr.flush();

// 2) flush produced records in the downstream
// TODO: this will actually block on all produced records across the tasks
Expand Down Expand Up @@ -287,7 +302,7 @@ public void close() {
throw exception;

try {
((ProcessorContextImpl) processorContext).stateManager().close(recordCollector.offsets());
stateMgr.close(recordCollector.offsets());
} catch (IOException e) {
throw new KafkaException("Error while closing the state manager in processor context", e);
}
Expand Down