Unable to append a new message to a topic in Camus.Getting "null record" error

[Ashish Dutt]

Hello all,

I am using Camus to insert messages into Kafka which then I will upload to HDFS. I'm using CDH5.4 with parcels and JSON file format for data

I am able to create a new topic and insert messages into it at one go but if I try to insert a new message or append a new message into the topic it fails giving me an error code: Java.lang.RunTimeException: Null record.

The camus.properties file is as follows

# Needed Camus properties, more cleanup to come

#

# Almost all properties have decent default properties. When in doubt, comment out the property.

#

# The job name.

camus.job.name=Camus Job

fs.defaultFS=hdfs://BDA01:8020

etl.record.writer.provider.class=com.linkedin.camus.etl.kafka.common.StringRecordWriterProvider

#etl.record.writer.provider.class=com.linkedin.camus.etl.kafka.common.AvroRecordWriterProvider

# final top-level data output directory, sub-directory will be dynamically created for each topic pulled

etl.destination.path=/user/kafka/

# HDFS location where you want to keep execution files, i.e. offsets, error logs, and count files

etl.execution.base.path=/user/kafka/exec

# where completed Camus job output directories are kept, usually a sub-dir in the base.path

etl.execution.history.path=/user/kafka/camus/exec/history

# Concrete implementation of the Encoder class to use (used by Kafka Audit, and thus optional for now)

#camus.message.encoder.class=com.linkedin.camus.etl.kafka.coders.DummyKafkaMessageEncoder

# Concrete implementation of the Decoder class to use.

# Out of the box options are:

#  com.linkedin.camus.etl.kafka.coders.JsonStringMessageDecoder - Reads JSON events, and tries to extract timestamp.

#  com.linkedin.camus.etl.kafka.coders.KafkaAvroMessageDecoder - Reads Avro events using a schema from a configured schema repository.

#  com.linkedin.camus.etl.kafka.coders.LatestSchemaKafkaAvroMessageDecoder - Same, but converts event to latest schema for current topic.

camus.message.decoder.class=com.linkedin.camus.etl.kafka.coders.JsonStringMessageDecoder

#camus.message.decoder.class=com.linkedin.camus.etl.kafka.coders.LatestSchemaKafkaAvroMessageDecoder

# Decoder class can also be set on a per topic basis.

#camus.message.decoder.class.<topic-name>=com.your.custom.MessageDecoder

# Used by avro-based Decoders (KafkaAvroMessageDecoder and LatestSchemaKafkaAvroMessageDecoder) to use as their schema registry.

# Out of the box options are:

# com.linkedin.camus.schemaregistry.FileSchemaRegistry

# com.linkedin.camus.schemaregistry.MemorySchemaRegistry

# com.linkedin.camus.schemaregistry.AvroRestSchemaRegistry

# com.linkedin.camus.example.schemaregistry.DummySchemaRegistry

kafka.message.coder.schema.registry.class=com.linkedin.camus.example.schemaregistry.DummySchemaRegistry

# Used by JsonStringMessageDecoder when extracting the timestamp

# Choose the field that holds the time stamp (default "timestamp")

camus.message.timestamp.field=time

# What format is the timestamp in? Out of the box options are:

# "unix" or "unix_seconds": The value will be read as a long containing the seconds since epoc

# "unix_milliseconds": The value will be read as a long containing the milliseconds since epoc

# "ISO-8601": Timestamps will be fed directly into org.joda.time.DateTime constructor, which reads ISO-8601

# All other values will be fed into the java.text.SimpleDateFormat constructor, which will be used to parse the timestamps

# Default is "[dd/MMM/yyyy:HH:mm:ss Z]"

#camus.message.timestamp.format=yyyy-MM-dd_HH:mm:ss

camus.message.timestamp.format=ISO-8601

# Used by the committer to arrange .avro files into a partitioned scheme. This will be the default partitioner for all

# topic that do not have a partitioner specified.

# Out of the box options are (for all options see the source for configuration options):

# com.linkedin.camus.etl.kafka.partitioner.HourlyPartitioner, groups files in hourly directories

# com.linkedin.camus.etl.kafka.partitioner.DailyPartitioner, groups files in daily directories

# com.linkedin.camus.etl.kafka.partitioner.TimeBasedPartitioner, groups files in very configurable directories

# com.linkedin.camus.etl.kafka.partitioner.DefaultPartitioner, like HourlyPartitioner but less configurable

# com.linkedin.camus.etl.kafka.partitioner.TopicGroupingPartitioner

#etl.partitioner.class=com.linkedin.camus.etl.kafka.partitioner.HourlyPartitioner

# Partitioners can also be set on a per-topic basis. (Note though that configuration is currently not per-topic.)

#etl.partitioner.class.<topic-name>=com.your.custom.CustomPartitioner

# all files in this dir will be added to the distributed cache and placed on the classpath for hadoop tasks

# hdfs.default.classpath.dir=

# max hadoop tasks to use, each task can pull multiple topic partitions

mapred.map.tasks=30

# max historical time that will be pulled from each partition based on event timestamp

kafka.max.pull.hrs=1

# events with a timestamp older than this will be discarded.

kafka.max.historical.days=3

# Max minutes for each mapper to pull messages (-1 means no limit)

kafka.max.pull.minutes.per.task=-1

# if whitelist has values, only whitelisted topic are pulled. Nothing on the blacklist is pulled

#kafka.blacklist.topics=

kafka.whitelist.topics=test

log4j.configuration=true

# Name of the client as seen by kafka

kafka.client.name=camus

# The Kafka brokers to connect to, format: kafka.brokers=host1:port,host2:port,host3:port

kafka.brokers=BDA01:9092,BDA02:9092, BDA03:9092, BDA04:9092

# Fetch request parameters:

#kafka.fetch.buffer.size=

#kafka.fetch.request.correlationid=

#kafka.fetch.request.max.wait=

#kafka.fetch.request.min.bytes=

#kafka.timeout.value=

#Stops the mapper from getting inundated with Decoder exceptions for the same topic

#Default value is set to 10

max.decoder.exceptions.to.print=5

#Controls the submitting of counts to Kafka

#Default value set to true

post.tracking.counts.to.kafka=true

#monitoring.event.class=class.that.generates.record.to.submit.counts.to.kafka

# everything below this point can be ignored for the time being, will provide more documentation down the road

##########################

etl.run.tracking.post=true

#kafka.monitor.tier=

#etl.counts.path=

kafka.monitor.time.granularity=10

#etl.hourly=hourly

#etdal.ily=daily

# Should we ignore events that cannot be decoded (exception thrown by MessageDecoder)?

# `false` will fail the job, `true` will silently drop the event.

etl.ignore.schema.errors=true

# configure output compression for deflate or snappy. Defaults to deflate

mapred.output.compress=false

#etl.output.codec=deflate

#etl.deflate.level=6

#etl.output.codec=snappy

etl.default.timezone=America/Los_Angeles

etl.output.file.time.partition.mins=60

etl.keep.count.files=false

etl.execution.history.max.of.quota=.8

# Configures a customer reporter which extends BaseReporter to send etl data

#etl.reporter.class

mapred.map.max.attempts=1

kafka.client.buffer.size=20971520

kafka.client.so.timeout=60000

#zookeeper.session.timeout=

#zookeeper.connection.timeout=

When I execute this command to append a new message to the topic I get this error in attached screenshot

hadoop jar /opt/camus/camus-example/target/camus-example-0.1.0-SNAPSHOT-shaded.jar com.linkedin.camus.etl.kafka.CamusJob -P /opt/camus/camus.properties

Please can anyone point me to the solution to how to fix it.

Thank you,

Ashish

[Rodolfo Mijangos]

Hello,   try changing 

camus.message.timestamp.field=time

to 

camus.message.timestamp.field=timestamp 

in your camus.properties 

[Ashish Dutt]

Many thanks Rodolfo, that did the trick. So now the conf.properties file reads as following

However, now when I try to insert a new message to a topic using the command 

/opt/cloudera/parcels/KAFKA/bin/kafka-console-producer --broker-list localhost:2181 --topic test

I type a message like 'Testing message:80'

I get the following error message

"received -1 when reading from channel. Socket has likely been closed"

Attached are the screenshots for the same

Sincerely,

Ashish Dutt

--
You received this message because you are subscribed to a topic in the Google Groups "Camus - Kafka ETL for Hadoop" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/camus_etl/hJFLtaC_P8Q/unsubscribe.
To unsubscribe from this group and all its topics, send an email to camus_etl+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.