flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-8005) Snapshotting FlinkKafkaProducer011 fails due to ClassLoader issues
Date Thu, 09 Nov 2017 13:57:00 GMT

    [ https://issues.apache.org/jira/browse/FLINK-8005?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16245681#comment-16245681
] 

ASF GitHub Bot commented on FLINK-8005:
---------------------------------------

Github user aljoscha commented on the issue:

    https://github.com/apache/flink/pull/4980
  
    Yes, but I think this is making an assumption about the internal implementation. If someone
changes that the test could break/not test the right thing anymore.


> Snapshotting FlinkKafkaProducer011 fails due to ClassLoader issues
> ------------------------------------------------------------------
>
>                 Key: FLINK-8005
>                 URL: https://issues.apache.org/jira/browse/FLINK-8005
>             Project: Flink
>          Issue Type: Bug
>          Components: Core, Kafka Connector, State Backends, Checkpointing
>    Affects Versions: 1.4.0
>            Reporter: Gary Yao
>            Assignee: Gary Yao
>            Priority: Blocker
>             Fix For: 1.4.0
>
>
> *Problem Description*
> Classes in the user code jar cannot be loaded by the snapshot thread’s context class
loader ({{AppClassLoader}}).
> For example, when creating instances of {{KafkaProducer}}, Strings are resolved to class
objects by Kafka.
> Find below an extract from {{ConfigDef.java}}:
> {code}
> case CLASS:
>     if (value instanceof Class)
>         return value;
>     else if (value instanceof String)
>         return Class.forName(trimmed, true, Utils.getContextOrKafkaClassLoader());
>     else
>         throw new ConfigException(name, value, "Expected a Class instance or class name.");
> {code}
> *Exception/Stacktrace*
> {noformat}
> Caused by: java.lang.Exception: Could not complete snapshot 1 for operator Source: Collection
Source -> Sink: kafka-sink-1510048188383 (1/1).
> 	at org.apache.flink.streaming.api.operators.AbstractStreamOperator.snapshotState(AbstractStreamOperator.java:379)
> 	at org.apache.flink.streaming.runtime.tasks.StreamTask$CheckpointingOperation.checkpointStreamOperator(StreamTask.java:1077)
> 	at org.apache.flink.streaming.runtime.tasks.StreamTask$CheckpointingOperation.executeCheckpointing(StreamTask.java:1026)
> 	at org.apache.flink.streaming.runtime.tasks.StreamTask.checkpointState(StreamTask.java:659)
> 	at org.apache.flink.streaming.runtime.tasks.StreamTask.performCheckpoint(StreamTask.java:595)
> 	at org.apache.flink.streaming.runtime.tasks.StreamTask.triggerCheckpoint(StreamTask.java:526)
> 	... 7 more
> Caused by: org.apache.kafka.common.config.ConfigException: Invalid value org.apache.kafka.common.serialization.ByteArraySerializer
for configuration key.serializer: Class org.apache.kafka.common.serialization.ByteArraySerializer
could not be found.
> 	at org.apache.kafka.common.config.ConfigDef.parseType(ConfigDef.java:715)
> 	at org.apache.kafka.common.config.ConfigDef.parseValue(ConfigDef.java:460)
> 	at org.apache.kafka.common.config.ConfigDef.parse(ConfigDef.java:453)
> 	at org.apache.kafka.common.config.AbstractConfig.<init>(AbstractConfig.java:62)
> 	at org.apache.kafka.common.config.AbstractConfig.<init>(AbstractConfig.java:75)
> 	at org.apache.kafka.clients.producer.ProducerConfig.<init>(ProducerConfig.java:360)
> 	at org.apache.kafka.clients.producer.KafkaProducer.<init>(KafkaProducer.java:288)
> 	at org.apache.flink.streaming.connectors.kafka.internal.FlinkKafkaProducer.<init>(FlinkKafkaProducer.java:114)
> 	at org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducer011.initProducer(FlinkKafkaProducer011.java:913)
> 	at org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducer011.initTransactionalProducer(FlinkKafkaProducer011.java:904)
> 	at org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducer011.createOrGetProducerFromPool(FlinkKafkaProducer011.java:637)
> 	at org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducer011.beginTransaction(FlinkKafkaProducer011.java:613)
> 	at org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducer011.beginTransaction(FlinkKafkaProducer011.java:94)
> 	at org.apache.flink.streaming.api.functions.sink.TwoPhaseCommitSinkFunction.beginTransactionInternal(TwoPhaseCommitSinkFunction.java:359)
> 	at org.apache.flink.streaming.api.functions.sink.TwoPhaseCommitSinkFunction.snapshotState(TwoPhaseCommitSinkFunction.java:294)
> 	at org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducer011.snapshotState(FlinkKafkaProducer011.java:756)
> 	at org.apache.flink.streaming.util.functions.StreamingFunctionUtils.trySnapshotFunctionState(StreamingFunctionUtils.java:118)
> 	at org.apache.flink.streaming.util.functions.StreamingFunctionUtils.snapshotFunctionState(StreamingFunctionUtils.java:99)
> 	at org.apache.flink.streaming.api.operators.AbstractUdfStreamOperator.snapshotState(AbstractUdfStreamOperator.java:90)
> 	at org.apache.flink.streaming.api.operators.AbstractStreamOperator.snapshotState(AbstractStreamOperator.java:357)
> 	... 12 more
> {noformat}
> *How to reproduce*
> Note that the problem only appears when a job is deployed on a cluster. 
> # Build Flink 1.4
> # Build test job https://github.com/GJL/flink-kafka011-producer-test with {{mvn -o clean
install -Pbuild-jar}}
> # Start job:
> {noformat}
> bin/flink run -c com.garyyao.StreamingJob /pathto/flink-kafka011-producer/target/flink-kafka011-producer-1.0-SNAPSHOT.jar
> {noformat}



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message