[
https://issues.apache.org/jira/browse/FLINK-3911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15283471#comment-15283471
]
Fabian Hueske commented on FLINK-3911:
--------------------------------------
I verified that between Flink 0.10.x and 1.0.x the Enum ordinals of
{{SORTED_GROUP_COMBINE}} and {{SORTED_GROUP_REDUCE}} shifted by one.
I assume that you submit a job with a 0.10.x client to a 1.0.x cluster.
How are you submitting the job?
You need to make sure that the client client has the same version as the
cluster you are submitting the job to.
> Sort operation before a group reduce doesn't seem to be implemented on 1.0.2
> ----------------------------------------------------------------------------
>
> Key: FLINK-3911
> URL: https://issues.apache.org/jira/browse/FLINK-3911
> Project: Flink
> Issue Type: Bug
> Affects Versions: 1.0.2
> Environment: Linux Ubuntu, standalone cluster
> Reporter: Patrice Freydiere
> Labels: newbie
>
> i have this piece of code:
> // group by id and sort on field order
> DataSet<Tuple2<Long, byte[]>> waysGeometry =
> joinedWaysWithPoints.groupBy(0).sortGroup(1, Order.ASCENDING)
> .reduceGroup(new
> GroupReduceFunction<Tuple4<Long, Integer, Double, Double>, Tuple2<Long,
> byte[]>>() {
> @Override
> public void
> reduce(Iterable<Tuple4<Long, Integer, Double, Double>> values,
>
> Collector<Tuple2<Long, byte[]>> out) throws Exception {
> long id = -1;
> and this exception when executing ;
> ava.lang.Exception: The data preparation for task 'GroupReduce (GroupReduce
> at constructOSMStreams(ProcessOSM.java:112))' , caused an error: Unrecognized
> driver strategy for GroupReduce driver: SORTED_GROUP_COMBINE
> at org.apache.flink.runtime.operators.BatchTask.run(BatchTask.java:456)
> at
> org.apache.flink.runtime.operators.BatchTask.invoke(BatchTask.java:345)
> at org.apache.flink.runtime.taskmanager.Task.run(Task.java:559)
> at java.lang.Thread.run(Thread.java:745)
> Caused by: java.lang.Exception: Unrecognized driver strategy for GroupReduce
> driver: SORTED_GROUP_COMBINE
> at
> org.apache.flink.runtime.operators.GroupReduceDriver.prepare(GroupReduceDriver.java:90)
> at org.apache.flink.runtime.operators.BatchTask.run(BatchTask.java:450)
> ... 3 more
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
