Skip to content

[vpj] Fix NorthGuard repush ZSTD dictionary routing #2942

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
sushantmane merged 3 commits into from
Jul 30, 2026
Merged

[vpj] Fix NorthGuard repush ZSTD dictionary routing #2942

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
93f668d
[vpj] Fix NorthGuard repush ZSTD dictionary routing
pthirun Jul 29, 2026
8738281
[vpj] Preserve pub-sub config for dictionary training
pthirun Jul 29, 2026
ca71159
[vpj] Cover dictionary consumer SSL branch
pthirun Jul 29, 2026
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
45 changes: 30 additions & 15 deletions clients/venice-push-job/src/main/java/com/linkedin/venice/hadoop/VenicePushJob.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
import static com.linkedin.venice.ConfigKeys.KAFKA_PRODUCER_REQUEST_TIMEOUT_MS;
import static com.linkedin.venice.ConfigKeys.KAFKA_PRODUCER_RETRIES_CONFIG;
import static com.linkedin.venice.ConfigKeys.MULTI_REGION;
import static com.linkedin.venice.ConfigKeys.PUBSUB_BROKER_ADDRESS;
import static com.linkedin.venice.ConfigKeys.VENICE_PARTITIONERS;
import static com.linkedin.venice.VeniceConstants.DEFAULT_SSL_FACTORY_CLASS_NAME;
import static com.linkedin.venice.status.BatchJobHeartbeatConfigs.HEARTBEAT_ENABLED_CONFIG;
Expand Down Expand Up @@ -866,13 +867,8 @@ public void run() {
if (pushJobSetting.isSourceKafka) {
if (pushJobSetting.sourceVersionCompressionStrategy == CompressionStrategy.ZSTD_WITH_DICT) {
LOGGER.info("Source version uses ZSTD_WITH_DICT. Fetching source dictionary.");
Properties kafkaConsumerProperties = new Properties();
if (pushJobSetting.enableSSL) {
kafkaConsumerProperties.putAll(this.sslProperties.get());
}
kafkaConsumerProperties.setProperty(KAFKA_BOOTSTRAP_SERVERS, pushJobSetting.repushSourcePubsubBroker);
ByteBuffer sourceDict = DictionaryUtils
.readDictionaryFromKafka(pushJobSetting.kafkaInputTopic, new VeniceProperties(kafkaConsumerProperties));
.readDictionaryFromKafka(pushJobSetting.kafkaInputTopic, getSourceDictionaryConsumerProperties());
if (sourceDict != null) {
pushJobSetting.sourceDictionary = ByteUtils.extractByteArray(sourceDict);
}
Expand Down Expand Up @@ -1666,12 +1662,35 @@ private Optional<ByteBuffer> getCompressionDictionary() throws VeniceException {
return Optional.of(emptyPushZstdDictionary.get());
}

private VeniceProperties getSourceDictionaryConsumerProperties() {
return getSourceDictionaryConsumerProperties(pushJobSetting.repushSourcePubsubBroker);
}

@VisibleForTesting
VeniceProperties getSourceDictionaryConsumerProperties(String sourcePubsubBroker) {
return buildSourceDictionaryConsumerProperties(
props,
pushJobSetting.enableSSL ? sslProperties.get() : new Properties(),
sourcePubsubBroker);
}

@VisibleForTesting
static VeniceProperties buildSourceDictionaryConsumerProperties(
VeniceProperties jobProperties,
Properties sslProperties,
String sourcePubsubBroker) {
Properties consumerProperties = jobProperties.toProperties();
consumerProperties.putAll(sslProperties);
consumerProperties.setProperty(PUBSUB_BROKER_ADDRESS, sourcePubsubBroker);
consumerProperties.setProperty(KAFKA_BOOTSTRAP_SERVERS, sourcePubsubBroker);
return new VeniceProperties(consumerProperties);
}

private ByteBuffer fetchOrBuildCompressionDictionary() throws VeniceException {
// Prepare the param builder, which can be used by different scenarios.
KafkaInputDictTrainer.ParamBuilder paramBuilder = new KafkaInputDictTrainer.ParamBuilder()
.setKeySchema(AvroCompatibilityHelper.toParsingForm(pushJobSetting.storeKeySchema))
.setNewKMESchemasFromController(pushJobSetting.newKmeSchemasFromController)
.setSslProperties(pushJobSetting.enableSSL ? sslProperties.get() : new Properties())
.setCompressionDictSize(
props.getInt(
COMPRESSION_DICTIONARY_SIZE_LIMIT,
Expand All @@ -1686,19 +1705,14 @@ private ByteBuffer fetchOrBuildCompressionDictionary() throws VeniceException {
LOGGER.info("Rebuild a new Zstd dictionary from the input topic: {}", pushJobSetting.kafkaInputTopic);
paramBuilder.setKafkaInputBroker(pushJobSetting.repushSourcePubsubBroker)
.setTopicName(pushJobSetting.kafkaInputTopic)
.setConsumerProperties(getSourceDictionaryConsumerProperties().toProperties())
.setSourceVersionCompressionStrategy(pushJobSetting.sourceKafkaInputVersionInfo.getCompressionStrategy());
KafkaInputDictTrainer dictTrainer = new KafkaInputDictTrainer(paramBuilder.build());
return ByteBuffer.wrap(dictTrainer.trainDict());
} else {
LOGGER.info("Reading Zstd dictionary from input topic: {}", pushJobSetting.kafkaInputTopic);
// set up ssl properties and kafka consumer properties
Properties kafkaConsumerProperties = new Properties();
if (pushJobSetting.enableSSL) {
kafkaConsumerProperties.putAll(this.sslProperties.get());
}
kafkaConsumerProperties.setProperty(KAFKA_BOOTSTRAP_SERVERS, pushJobSetting.repushSourcePubsubBroker);
return DictionaryUtils
.readDictionaryFromKafka(pushJobSetting.kafkaInputTopic, new VeniceProperties(kafkaConsumerProperties));
.readDictionaryFromKafka(pushJobSetting.kafkaInputTopic, getSourceDictionaryConsumerProperties());
}
}
LOGGER.info(
Expand Down Expand Up @@ -1740,8 +1754,9 @@ private ByteBuffer fetchOrBuildCompressionDictionary() throws VeniceException {
"Rebuild a new Zstd dictionary from the source topic: {} in Kafka: {}",
sourceTopicName,
sourceKafkaUrl);
paramBuilder.setKafkaInputBroker(repushInfoResponse.getRepushInfo().getKafkaBrokerUrl())
paramBuilder.setKafkaInputBroker(sourceKafkaUrl)
.setTopicName(sourceTopicName)
.setConsumerProperties(getSourceDictionaryConsumerProperties(sourceKafkaUrl).toProperties())
.setSourceVersionCompressionStrategy(
repushInfoResponse.getRepushInfo().getVersion().getCompressionStrategy());
KafkaInputDictTrainer dictTrainer = new KafkaInputDictTrainer(paramBuilder.build());
Expand Down
12 changes: 6 additions & 6 deletions ...-push-job/src/main/java/com/linkedin/venice/hadoop/input/kafka/KafkaInputDictTrainer.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,7 +50,7 @@ public static class Param {
private final String kafkaInputBroker;
private final String topicName;
private final String keySchema;
private final Properties sslProperties;
private final Properties consumerProperties;
private final int compressionDictSize;
private final int dictSampleSize;
private final CompressionStrategy sourceVersionCompressionStrategy;
Expand All @@ -62,7 +62,7 @@ public static class Param {
this.kafkaInputBroker = builder.kafkaInputBroker;
this.topicName = builder.topicName;
this.keySchema = builder.keySchema;
this.sslProperties = builder.sslProperties;
this.consumerProperties = builder.consumerProperties;
this.compressionDictSize = builder.compressionDictSize;
this.dictSampleSize = builder.dictSampleSize;
this.sourceVersionCompressionStrategy = builder.sourceVersionCompressionStrategy;
Expand All @@ -75,7 +75,7 @@ public static class ParamBuilder {
private String kafkaInputBroker;
private String topicName;
private String keySchema;
private Properties sslProperties;
private Properties consumerProperties;
private int compressionDictSize;
private int dictSampleSize;
private CompressionStrategy sourceVersionCompressionStrategy;
Expand All @@ -97,8 +97,8 @@ public ParamBuilder setKeySchema(String keySchema) {
return this;
}

public ParamBuilder setSslProperties(Properties sslProperties) {
this.sslProperties = sslProperties;
public ParamBuilder setConsumerProperties(Properties consumerProperties) {
this.consumerProperties = consumerProperties;
return this;
}

Expand Down Expand Up @@ -166,11 +166,11 @@ protected KafkaInputDictTrainer(
this.trainerSupplier = trainerSupplier;
this.sourceVersionCompressionStrategy = param.sourceVersionCompressionStrategy;
Properties properties = new Properties();
properties.putAll(param.consumerProperties);
properties.setProperty(VENICE_REPUSH_SOURCE_PUBSUB_BROKER, param.kafkaInputBroker);
properties.setProperty(KAFKA_INPUT_TOPIC, param.topicName);
properties.setProperty(KAFKA_SOURCE_KEY_SCHEMA_STRING_PROP, param.keySchema);
this.sourceTopicName = param.topicName;
properties.putAll(param.sslProperties);
properties.setProperty(COMPRESSION_DICTIONARY_SIZE_LIMIT, Integer.toString(param.compressionDictSize));
properties.setProperty(COMPRESSION_DICTIONARY_SAMPLE_SIZE, Integer.toString(param.dictSampleSize));
properties
Expand Down
57 changes: 57 additions & 0 deletions ...nts/venice-push-job/src/test/java/com/linkedin/venice/hadoop/VenicePushJobRepushTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,10 @@
package com.linkedin.venice.hadoop;

import static com.linkedin.venice.ConfigKeys.KAFKA_BOOTSTRAP_SERVERS;
import static com.linkedin.venice.ConfigKeys.PUBSUB_BROKER_ADDRESS;
import static com.linkedin.venice.ConfigKeys.PUBSUB_CONSUMER_ADAPTER_FACTORY_CLASS;
import static com.linkedin.venice.ConfigKeys.PUBSUB_SECURITY_PROTOCOL;
import static com.linkedin.venice.ConfigKeys.PUBSUB_TYPE_ID_TO_POSITION_CLASS_NAME_MAP;
import static com.linkedin.venice.vpj.VenicePushJobConstants.ALLOW_REGULAR_PUSH_WITH_TTL_REPUSH;
import static com.linkedin.venice.vpj.VenicePushJobConstants.COMPLIANCE_PUSH;
import static com.linkedin.venice.vpj.VenicePushJobConstants.KAFKA_INPUT_MAX_RECORDS_PER_MAPPER;
Expand All @@ -26,6 +31,7 @@
import com.linkedin.venice.meta.StoreInfo;
import com.linkedin.venice.meta.Version;
import com.linkedin.venice.utils.Time;
import com.linkedin.venice.utils.VeniceProperties;
import java.util.HashMap;
import java.util.Map;
import java.util.Properties;
Expand All @@ -42,6 +48,57 @@
*/

public class VenicePushJobRepushTest extends VenicePushJobTestBase {
@Test
public void testSourceDictionaryConsumerPropertiesRetainPubSubConfigAndOverrideBrokers() {
Properties jobProperties = new Properties();
jobProperties.setProperty(
PUBSUB_CONSUMER_ADAPTER_FACTORY_CLASS,
"com.linkedin.venice.pubsub.adapter.xinfra.consumer.XcConsumerAdapterFactory");
jobProperties.setProperty(
PUBSUB_TYPE_ID_TO_POSITION_CLASS_NAME_MAP,
"1:com.linkedin.venice.pubsub.adapter.xinfra.XinfraPosition");
jobProperties.setProperty("xc.pubsub.broker.url.to.region.name.map", "northguard:ei4");
jobProperties.setProperty(PUBSUB_BROKER_ADDRESS, "destination-broker");
jobProperties.setProperty(KAFKA_BOOTSTRAP_SERVERS, "legacy-broker");

try (VenicePushJob pushJob = getSpyVenicePushJob(jobProperties, null)) {
VeniceProperties consumerProperties = pushJob.getSourceDictionaryConsumerProperties("source-broker");

assertEquals(
consumerProperties.getString(PUBSUB_CONSUMER_ADAPTER_FACTORY_CLASS),
"com.linkedin.venice.pubsub.adapter.xinfra.consumer.XcConsumerAdapterFactory");
assertEquals(
consumerProperties.getString(PUBSUB_TYPE_ID_TO_POSITION_CLASS_NAME_MAP),
"1:com.linkedin.venice.pubsub.adapter.xinfra.XinfraPosition");
assertEquals(consumerProperties.getString("xc.pubsub.broker.url.to.region.name.map"), "northguard:ei4");
assertEquals(consumerProperties.getString(PUBSUB_BROKER_ADDRESS), "source-broker");
assertEquals(consumerProperties.getString(KAFKA_BOOTSTRAP_SERVERS), "source-broker");
assertEquals(
jobProperties.getProperty(PUBSUB_BROKER_ADDRESS),
"destination-broker",
"Building consumer properties must not mutate the job properties");
}
}

@Test
public void testSourceDictionaryConsumerPropertiesApplySslOverrides() {
Properties jobProperties = new Properties();
jobProperties.setProperty(PUBSUB_SECURITY_PROTOCOL, "PLAINTEXT");
jobProperties.setProperty("ssl.keystore.location", "stale-keystore");
jobProperties.setProperty("xc.tls.key.store.type", "PKCS12");

Properties sslProperties = new Properties();
sslProperties.setProperty(PUBSUB_SECURITY_PROTOCOL, "SSL");
sslProperties.setProperty("ssl.keystore.location", "credential-keystore");

VeniceProperties consumerProperties = VenicePushJob
.buildSourceDictionaryConsumerProperties(new VeniceProperties(jobProperties), sslProperties, "source-broker");

assertEquals(consumerProperties.getString(PUBSUB_SECURITY_PROTOCOL), "SSL");
assertEquals(consumerProperties.getString("ssl.keystore.location"), "credential-keystore");
assertEquals(consumerProperties.getString("xc.tls.key.store.type"), "PKCS12");
}

@Test(expectedExceptions = VeniceException.class, expectedExceptionsMessageRegExp = ".*Repush with TTL is only supported while using Kafka Input Format.*")
public void testRepushTTLJobWithNonKafkaInput() {
Properties repushProps = new Properties();
Expand Down
64 changes: 59 additions & 5 deletions ...h-job/src/test/java/com/linkedin/venice/hadoop/input/kafka/TestKafkaInputDictTrainer.java
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,12 +1,22 @@
package com.linkedin.venice.hadoop.input.kafka;

import static com.linkedin.venice.ConfigKeys.KAFKA_BOOTSTRAP_SERVERS;
import static com.linkedin.venice.ConfigKeys.PUBSUB_BROKER_ADDRESS;
import static com.linkedin.venice.ConfigKeys.PUBSUB_CONSUMER_ADAPTER_FACTORY_CLASS;
import static com.linkedin.venice.ConfigKeys.PUBSUB_SECURITY_PROTOCOL;
import static com.linkedin.venice.ConfigKeys.PUBSUB_TYPE_ID_TO_POSITION_CLASS_NAME_MAP;
import static com.linkedin.venice.vpj.VenicePushJobConstants.KAFKA_INPUT_TOPIC;
import static org.mockito.ArgumentMatchers.any;
import static org.mockito.ArgumentMatchers.eq;
import static org.mockito.Mockito.doAnswer;
import static org.mockito.Mockito.doReturn;
import static org.mockito.Mockito.mock;
import static org.mockito.Mockito.never;
import static org.mockito.Mockito.verify;
import static org.testng.Assert.assertEquals;
import static org.testng.Assert.assertFalse;
import static org.testng.Assert.assertTrue;
import static org.testng.Assert.fail;

import com.github.luben.zstd.ZstdDictTrainer;
import com.linkedin.venice.compression.CompressionStrategy;
Expand Down Expand Up @@ -37,6 +47,7 @@
import org.apache.avro.Schema;
import org.apache.hadoop.mapred.InputSplit;
import org.apache.hadoop.mapred.RecordReader;
import org.mockito.ArgumentCaptor;
import org.testng.annotations.Test;


Expand All @@ -52,6 +63,13 @@ private KafkaInputDictTrainer.Param getParam(int sampleSize) {
}

private KafkaInputDictTrainer.Param getParam(int sampleSize, CompressionStrategy sourceVersionCompressionStrategy) {
return getParam(sampleSize, sourceVersionCompressionStrategy, new Properties());
}

private KafkaInputDictTrainer.Param getParam(
int sampleSize,
CompressionStrategy sourceVersionCompressionStrategy,
Properties consumerProperties) {
Map<Integer, Schema> allSchemas = Utils.getAllSchemasFromResources(AvroProtocolDefinition.KAFKA_MESSAGE_ENVELOPE);
Map<Integer, String> allSchemaStr =
allSchemas.entrySet().stream().collect(Collectors.toMap(Map.Entry::getKey, e -> e.getValue().toString()));
Expand All @@ -60,14 +78,14 @@ private KafkaInputDictTrainer.Param getParam(int sampleSize, CompressionStrategy
.setKeySchema("\"string\"")
.setCompressionDictSize(900 * 1024)
.setDictSampleSize(sampleSize)
.setSslProperties(new Properties())
.setConsumerProperties(consumerProperties)
.setSourceVersionCompressionStrategy(sourceVersionCompressionStrategy)
.setNewKMESchemasFromController(allSchemaStr)
.build();
}

@Test(expectedExceptions = VeniceException.class, expectedExceptionsMessageRegExp = "No record.*")
public void testEmptyTopic() throws IOException {
@Test
public void testConsumerPropertiesPropagateToEmptyTopicTraining() throws IOException {
KafkaInputFormat mockFormat = mock(KafkaInputFormat.class);
PubSubTopicPartition topicPartition =
new PubSubTopicPartitionImpl(PUB_SUB_TOPIC_REPOSITORY.getTopic("test_topic"), 0);
Expand All @@ -79,12 +97,48 @@ public void testEmptyTopic() throws IOException {
doReturn(false).when(mockRecordReader).next(any(), any());
doReturn(mockRecordReader).when(mockFormat).getRecordReader(any(), any(), any(), any());

Properties consumerProperties = new Properties();
consumerProperties.setProperty(
PUBSUB_CONSUMER_ADAPTER_FACTORY_CLASS,
"com.linkedin.venice.pubsub.adapter.xinfra.consumer.XcConsumerAdapterFactory");
consumerProperties.setProperty(
PUBSUB_TYPE_ID_TO_POSITION_CLASS_NAME_MAP,
"1:com.linkedin.venice.pubsub.adapter.xinfra.XinfraPosition");
consumerProperties.setProperty("xc.pubsub.broker.url.to.region.name.map", "northguard:ei4");
consumerProperties.setProperty(PUBSUB_BROKER_ADDRESS, "test_url");
consumerProperties.setProperty(KAFKA_BOOTSTRAP_SERVERS, "test_url");
consumerProperties.setProperty(PUBSUB_SECURITY_PROTOCOL, "SSL");
consumerProperties.setProperty("ssl.keystore.location", "credential-keystore");

KafkaInputDictTrainer trainer = new KafkaInputDictTrainer(
mockFormat,
Optional.empty(),
getParam(100),
getParam(100, CompressionStrategy.NO_OP, consumerProperties),
getCompressorBuilder(new NoopCompressor()));
trainer.trainDict(Optional.of(mock(PubSubConsumerAdapter.class)));
try {
trainer.trainDict(Optional.of(mock(PubSubConsumerAdapter.class)));
fail("Expected training on an empty topic to fail");
} catch (VeniceException e) {
assertTrue(e.getMessage().startsWith("No record"));
}

ArgumentCaptor<VeniceProperties> consumerPropertiesCaptor = ArgumentCaptor.forClass(VeniceProperties.class);
verify(mockFormat).getSplits(consumerPropertiesCaptor.capture());
VeniceProperties actualConsumerProperties = consumerPropertiesCaptor.getValue();
assertEquals(
actualConsumerProperties.getString(PUBSUB_CONSUMER_ADAPTER_FACTORY_CLASS),
"com.linkedin.venice.pubsub.adapter.xinfra.consumer.XcConsumerAdapterFactory");
assertEquals(
actualConsumerProperties.getString(PUBSUB_TYPE_ID_TO_POSITION_CLASS_NAME_MAP),
"1:com.linkedin.venice.pubsub.adapter.xinfra.XinfraPosition");
assertEquals(actualConsumerProperties.getString("xc.pubsub.broker.url.to.region.name.map"), "northguard:ei4");
assertEquals(actualConsumerProperties.getString(PUBSUB_BROKER_ADDRESS), "test_url");
assertEquals(actualConsumerProperties.getString(KAFKA_BOOTSTRAP_SERVERS), "test_url");
assertEquals(actualConsumerProperties.getString(PUBSUB_SECURITY_PROTOCOL), "SSL");
assertEquals(actualConsumerProperties.getString("ssl.keystore.location"), "credential-keystore");
assertFalse(
consumerProperties.containsKey(KAFKA_INPUT_TOPIC),
"Building trainer properties must not mutate the supplied consumer properties");
}

interface ResettableRecordReader<K, V> extends RecordReader<K, V> {
Expand Down