[jira] [Comment Edited] (FLINK-11654) ProducerFencedExceptions from Kafka in EXACTLY_ONCE mode due to identical transactional IDs in multiple jobs

["Jürgen Kreileder (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-11654?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16773458#comment-16773458 ] 

Jürgen Kreileder edited comment on FLINK-11654 at 2/20/19 10:44 PM:
--------------------------------------------------------------------

I could probably provide a fix and test case for this if somebody can tell me if adding the JobID to the prefix is actually the right way of fixing it (something like [https://github.com/jkreileder/flink/commit/81100d07e03e3ec107e389ab35c59bd7cc9a6378]).

JobID is a 16-byte random value, so there's no need to add something like a cluster ID. But is JobID the right concept here?

 


was (Author: jkreileder):
I could probably provide a fix and test case for this if somebody can tell me if adding the JobID to the prefix is actually the right way of fixing it (something like [https://github.com/jkreileder/flink/commit/81100d07e03e3ec107e389ab35c59bd7cc9a6378]).

JobID is a 16-bit random value, so there's no need to add something like a cluster ID. But is JobID the right concept here?

 

> ProducerFencedExceptions from Kafka in EXACTLY_ONCE mode due to identical transactional IDs in multiple jobs
> ------------------------------------------------------------------------------------------------------------
>
>                 Key: FLINK-11654
>                 URL: https://issues.apache.org/jira/browse/FLINK-11654
>             Project: Flink
>          Issue Type: Bug
>          Components: Kafka Connector
>    Affects Versions: 1.7.1
>            Reporter: Jürgen Kreileder
>            Priority: Blocker
>             Fix For: 1.8.0
>
>
> We run multiple jobs on a cluster which write a lot to the same Kafka topic from identically named sinks. When EXACTLY_ONCE semantic is enabled for the KafkaProducers we run into a lot of ProducerFencedExceptions and all jobs go into a restart cycle.
> Example exception from the Kafka log:
>  
> {code:java}
> [2019-02-18 18:05:28,485] ERROR [ReplicaManager broker=1] Error processing append operation on partition finding-commands-dev-1-0 (kafka.server.ReplicaManager)
> org.apache.kafka.common.errors.ProducerFencedException: Producer's epoch is no longer valid. There is probably another producer with a newer epoch. 483 (request epoch), 484 (server epoch)
> {code}
> The reason for this is the way FlinkKafkaProducer initializes the TransactionalIdsGenerator:
> The IDs are only guaranteed to be unique for a single Job. But they can clash between different Jobs (and Clusters).
>  
>  
> {code:java}
> --- a/flink-connectors/flink-connector-kafka/src/main/java/org/apache/flink/streaming/connectors/kafka/FlinkKafkaProducer.java
> +++ b/flink-connectors/flink-connector-kafka/src/main/java/org/apache/flink/streaming/connectors/kafka/FlinkKafkaProducer.java
> @@ -819,6 +819,7 @@ public class FlinkKafkaProducer<IN>
>                 nextTransactionalIdHintState = context.getOperatorStateStore().getUnionListState(
>                         NEXT_TRANSACTIONAL_ID_HINT_DESCRIPTOR);
>                 transactionalIdsGenerator = new TransactionalIdsGenerator(
> + // the prefix probably should include job id and maybe cluster id
>                         getRuntimeContext().getTaskName() + "-" + ((StreamingRuntimeContext) getRuntimeContext()).getOperatorUniqueID(),
>                         getRuntimeContext().getIndexOfThisSubtask(),
>                         getRuntimeContext().getNumberOfParallelSubtasks(),{code}
>  
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)