GetNewApplicationRequestProto cannot be cast to com.google.protobuf Message

[tangsho...@gmail.com]

When i use JanusGraph2.0 and spark2.11 on yarn to computer e.g.:

gremlin> graph = GraphFactory.open('conf/hadoop-graph/hadoop-load.properties')
==>hadoopgraph[gryoinputformat->nulloutputformat]
gremlin> blvp = BulkLoaderVertexProgram.build().writeGraph('conf/janusgraph-cassandra.properties').create(graph)
==>BulkLoaderVertexProgram[bulkLoader=IncrementalBulkLoader,vertexIdProperty=bulkLoader.vertex.id,userSuppliedIds=false,keepOriginalIds=true,batchSize=0]
gremlin> graph.compute(SparkGraphComputer).program(blvp).submit().get()

it throws an exception:

08:57:03 ERROR org.apache.spark.SparkContext  - Error initializing SparkContext.
java.lang.ClassCastException: org.apache.hadoop.yarn.proto.YarnServiceProtos$GetNewApplicationRequestProto cannot be cast to com.google.protobuf.Message
        at org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.invoke(ProtobufRpcEngine.java:226)
        at com.sun.proxy.$Proxy23.getNewApplication(Unknown Source)

Can anyone help me solve this problem?

[Jason Plurad]

Check the version compatibility matrix. JanusGraph 0.2.0 is not compatible with Spark 2.11.
http://docs.janusgraph.org/latest/version-compat.html

[tangsho...@gmail.com]

eh eh, what a stupid mistake! thx！

在 2018年4月9日星期一 UTC+8上午10:04:26，Jason Plurad写道：

[tangsho...@gmail.com]

While i change to spark 1.6.3 and hadoop 2.6.5, this problem still exists！look forward to your help!

在 2018年4月9日星期一 UTC+8上午10:04:26，Jason Plurad写道：

Check the version compatibility matrix. JanusGraph 0.2.0 is not compatible with Spark 2.11.

[jiang xingkai]

Hi,have you solved the problem?

在 2018年4月9日星期一 UTC+8下午4:09:07，tangsho...@gmail.com写道：

[Debasish Kanhar]

Use JanusGraph 0.2.1 with Spark 1.6.1. I've been able to make it work on my end with Cassandra as storage backend and Elasticearch as indexing backend.

There was a bug with JanusGraph 0.2.0 as it didn't contain multiple Hadoop jars in its dist. Hence if we try to run on Spark cluster, the job will fail. The same was rectified in v 0.2.1.