Hi,
I am using the flink training exercise TaxiRide [1] to execute a
stream count of events. On the cluster and on my local machine I am
receiving the message that joda.Time cannot be serialized "class
org.joda.time.LocalDateTime is not a valid POJO type". However it is
starting the job on the cluster, but not in my local machine. So I
searched in the internet and it is requested to register the jodaTime
class on the environment[2]. I did like this:
env.getConfig().registerTypeWithKryoSerializer(DateTime.class,
AvroKryoSerializerUtils.JodaDateTimeSerializer.class);
env.getConfig().registerTypeWithKryoSerializer(LocalDate.class,
AvroKryoSerializerUtils.JodaLocalDateSerializer.class);
env.getConfig().registerTypeWithKryoSerializer(LocalTime.class,
AvroKryoSerializerUtils.JodaLocalTimeSerializer.class);
and I added the joda and avro dependency on the pom.xml:
<dependency>
<groupId>joda-time</groupId>
<artifactId>joda-time</artifactId>
</dependency>
<dependency>
<groupId>org.apache.flink</groupId>
<artifactId>flink-avro</artifactId>
<version>${project.version}</version>
</dependency>
I also tested using addDefaultKryoSerializer but I got the same error.
For some reason, it is still not working. Does anyone have some hint
of what could be happening?
Thanks! Felipe
[1]
https://github.com/ververica/flink-training-exercises/blob/master/src/main/java/com/ververica/flinktraining/exercises/datastream_java/datatypes/TaxiRide.java
[2]
https://ci.apache.org/projects/flink/flink-docs-stable/dev/custom_serializers.html
--
-- Felipe Gutierrez
-- skype: felipe.o.gutierrez
-- https://felipeogutierrez.blogspot.com
